Words - An Integrational Approach [1 ed.] 1138868345, 9781138868342

Table of contents :
Cover
WORDS: An Integrational Approach
Copyright
Contents
Preface
Abbreviations
1 Orientation: The Word of Linguistic Theory
2 Methodology: The Word of the Layperson
3 What do Lay Speakers Say About Words?
4 Words and Linguistic Meaning
5 Parts of Speech and Grammar
6 Folk Characteristics of Words
7 Reorientation: the Integration of Speech and Writing
Bibliography
Index

Citation preview

WORDS An Integrational Approach

WORDS An Integrational Approach

Hayley G. Davis

This edition published 2011 by Routledge Routledge Taylor & Francis Group 711 Third Avenue New York, NY 10017

Routledge Taylor & Francis Group 2 Park Square, Milton Park Abingdon, Oxon OX14 4RN

First Published in 2001 by Curzon Press Richmond, Surrey # 2001 Hayley G. Davis Typeset in Times by LaserScript Ltd, Mitcham, Surrey All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record of this book is available from the British Library Library of Congress Cataloguing in Publication Data A catalogue record for this book has been requested ISBN 0±7007±1376±X

To Mum and Dad

Contents

Preface Abbreviations

ix xi

1 Orientation: The Word of Linguistic Theory

1

2 Methodology: The Word of the Layperson

21

3 What do Lay Speakers Say About Words?

49

4 Words and Linguistic Meaning

87

5 Parts of Speech and Grammar

117

6 Folk Characteristics of Words

147

7 Reorientation: The Integration of Speech and Writing

191

Bibliography Index

207 215

vii

Preface

This book coincides with the establishment of The International Association for the Integrational Study of Language and Communication (IAISLC), which was founded in 1998 for the purpose of promoting an integrational approach to the study of language and communication. During July 1998 eight of us ± Daniel Davis, Roy Harris, Chris Hutton, Nigel Love, Talbot Taylor, Michael Toolan, George Wolf and myself ± accompanied by partners and offspring, descended upon the French town of Chablis in order to discuss the setting up of an integrationist association. We are all, in various ways, committed to reorienting the study of language by taking into serious consideration the perspective on linguistic matters taken by lay speakers themselves. We all feel, to various degrees, that linguistics is in need of such a reorientation as a response to the now inescapable conclusion that traditional linguistic theory, with its focus on revealing `the facts of language in general' cannot handle the necessary indeterminacy of what is said and understood. Integrationists insist that in order to understand language we must consider how it is actually used and understood by individual language users. In order to show how this approach is viable, I have focused on the uses to which English speakers on the one hand, and linguistic theorists on the other, put the word word. This type of analysis also serves as an example of the more general contrast between how an `average' person and how a professional linguist treats metalinguistic remarks. To investigate the views of lay speakers I have employed an ask-the-speaker methodology, using ix

Preface

®eldwork and interview techniques. This study is the ®rst attempt to make such an analysis and to draw the relevant theoretical conclusions concerning the status of the linguistic units in question. Chapter One considers the way in which the concept of the word is treated in linguistic theory. The focus in Chapter Two is on the `word' of the ordinary speaker. In Chapters Three and Four I discuss the different ways in which linguists and laypeople discuss the issues of `linguistic form' and `linguistic meaning'. Chapter Five discusses, and where appropriate, compares the way linguists and laypeople talk about `parts of speech' and grammar. In Chapter Six, I look at the different `value judgements' which may be assigned to particular words. Chapter Seven looks at the in¯uence of literacy on both professional linguists' and laypeople's metalinguistic judgements. I thank all those who agreed to be my informants. Your names have been changed but I'm sure that you will recognise yourselves. I should also like to thank Ron, the landlord of the Happy Man, who allowed me to interview some of his drinkers. I would also like to thank the members of IAISLC named above who have, in the past, commented on parts, or in the cases of Roy Harris, Nigel Love, Talbot Taylor and Michael Toolan, on the whole of this work. A special thanks goes to Roy Harris, without whom this book, which began its life as a D. Phil. thesis, would never have been realized. I am also grateful to Chris Wells for his helpful comments on early drafts of this work. My love and appreciation also extends to Chris Baldick and Bethany for everything. I also thank my other children Joshua, ZoeÈ and Ella for their patience: perhaps they will read this one day and understand why I was asking them such `silly' questions.

x

Abbreviations

BB CLG COD OED PG PI PR TLP

The Blue and Brown Books, Wittgenstein 1958 Course de linguistique geÂneÂrale, Saussure 1922 Concise Oxford Dictionary Oxford English Dictionary Philosophical Grammar, Wittgenstein 1974 Philosophical Investigations, Wittgenstein 1958 Philosophical Remarks, Wittgenstein 1975 Tractatus Logico-philosphicus, Wittgenstein 1972

xi

1 Orientation: The Word of Linguistic Theory

The main themes and objectives of this book are to reorient the study of language by taking into serious consideration the perspective on linguistic matters taken by lay speakers themselves. The incorporation of such a `folk theoretical' methodology would bring linguistics into line with the trend in related social science disciplines (e.g. psychology, anthropology, and sociology) to assign a central methodological importance to the irreducibly interpreted and negotiated character of human behaviour and interactional events. Linguistics, in particular, is in need of such a reorientation as a response to the now inescapable conclusion that traditional linguistic theory, with its focus on revealing `the facts of language in general' cannot handle the necessary indeterminacy of what is said and understood. Furthermore, incorporating a folk-theoretical approach, it will be argued, is a necessary ampli®cation of language study since it is by the expression of their own re¯exive understandings that lay speakers are able to impose regularities and constraints upon language use: regularities and constraints which cannot therefore be explained without reference to those understandings. This proposed reorientation, involving an ever shifting perspective, therefore demands that we reconceive how we approach metalinguistic and metacommunicative issues, since `language', `word' and `meaning' etc. are not super-concepts pre-existing their `ordinary' uses by `ordinary' speakers (Wittgenstein 1951). Since the concept of the word is central both to linguistic theory and to the vocabulary of a native English speaker, the 1

Orientation: The Word of Linguistic Theory

focus of this book will be on the uses to which English speakers, on the one hand, and linguistic theorists, on the other, put the word word. The analysis will therefore serve as an example of the more general contrast between how an `average' person and how a professional linguist treats metalinguistic remarks (see Taylor 1992). To investigate the views of lay speakers1, an ask-thespeaker methodology is employed, using ®eldwork and interview techniques. This study is the ®rst attempt to make such an analysis and to draw the relevant theoretical conclusions concerning the status of the linguistic units in question. The Word of Linguistic Theory The questions `what is a word?', or `how are we to de®ne a word?' have, in the past been a matter of serious concern to many linguistic theorists. However, such concerns rarely, if at all, feature in the talk of ordinary people. Even children could sensibly be expected to de®ne and distinguish between sentences and clauses at school2, but would rarely ®nd themselves in situations appropriate to answering such metalinguistic questions as `what is a word?', or `can you de®ne a word?'. In short, ordinary speakers who are familiar with the concept `word' need to enquire into its uses or de®nitions only if playing a game with rigid (non-negotiable) rules like Scrabble (`em' is a word ± it's in Webster's New World Dictionary, de®ned as the thirteenth letter of the alphabet) or performing an activity such as a word-count for an assessed essay (do we count hyphenated words such as `non-literary' as one word or two?). This is why ordinary speakers might well be baf¯ed if they knew some of the activities linguists got up to: de®ning a word in order to contribute to the science of language. 1 The somewhat problematic terms lay speaker and ordinary language user are employed simply to contrast the speakers with professional and academic linguists. The informant pro®les on 44ff should acknowledge the complete diversity in all my `subjects'. 2 The proposed syllabus 1998 for the National Curriculum (English) states that at Key Stage 2 (7±11 years), `students should be taught the features of sentence types including statements, questions, commands'.

2

Orientation: The Word of Linguistic Theory

The Speech Circuit Model of Communication The procedures of linguists thus form an exception to the above word-based activities performed by ordinary speakers. The term word for linguists, whether they like it or not, has become a technical term of central importance in modern linguistic analysis partly because of the model of communication they have inherited. This model assumes that talking involves uttering units which have the `same' interpretation and representation for all speakers of the `same' language, as succinctly described by Saussure's `speech circuit': The starting point of the circuit is in the brain of one individual, for instance A, where facts of consciousness which we shall call concepts are associated with representations of linguistic signs or sound patterns by means of which they may be expressed. Let us suppose that a given concept triggers in the brain a corresponding sound pattern. This is an entirely psychological phenomenon, followed in turn by a physiological process: the brain transmits to the organs of phonotation an impulse corresponding to the pattern, then sound waves are sent from A's mouth to B's ear: a purely physical process (Saussure 1922: 28). That Saussure's model of communication is today considered basically correct can be seen in more contemporary discussions of language and communication: Something in Harry's brain that we might as well call a `thought' results in movements of his vocal tract (lungs, vocal cords, tongue, jaw, and lips), which in turn create a sound wave that is transmitted through the air. This sound wave, striking Sam's ear, results in Sam's having the same `thought' (or a similar one) in his brain (Jackendoff 1993: 3). Knowledge of a language is thus explained by linguists in terms of knowing which forms are correlated with which meanings, 3

Orientation: The Word of Linguistic Theory

and how the units combine into larger elements to form utterances by means of ®xed `rules': in order for us to speak and understand novel sentences, we have to store in our heads not just the words of our language but also the patterns of sentences possible in our language. These patterns, in turn, describe not just patterns of words but also patterns of patterns. Linguists refer to these patterns as the rules of language stored in memory; they refer to the complete collection of rules as the mental grammar of the language or grammar for short (Jackendoff 1993: 14). The rules have to be ®xed under this model in order for A and B to arrive at the `same' interpretation of the utterance. It would be a sorry state of affairs if B regularly couldn't distinguish between such verb phrases as `running up' as in `David was running up a bill' and `David was running up a hill' or between `eating up' as in `Dorothy was eating up a tree' and `Dorothy was eating up a cake'. In some models of language adopted by contemporary linguists, it is therefore the `word' and not the sentence or utterance which has the key role, since it is claimed that if we (and our brains) are to be able to characterize the patterns of phrases and sentences, we need an additional layer in which the basic units of analysis are parts of speech, and in which they are combined into phrases and sentences (ibid.: 67). For some, even, the `principal task of linguistics is to investigate and describe the ways in which words can be combined and manipulated to convey meanings' (Brown 1984: 9). The difference between the various linguistic theories are mainly concerned with the ontological status of such units3. Thus according to Saussure and subsequent linguists, communication 3 i.e. whether words are, within orthodox linguistic theories, (a) `real' objects inherent in the language, (b) abstract entities in speakers' minds or (c) merely cultural constructs, a result of (linguistic) theorising.

4

Orientation: The Word of Linguistic Theory

is a form of telementation4 or thought transference conducted via the medium of spoken linguistic signs. This view necessarily leads to the assumption that linguistic units have determinate forms and meanings. To deny this would raise too many problems. For one, how else could communication work? It appears a matter of common sense that we can recognise and say the `same' words, thus ensuring successful communication. If John says to Norma shake my hand and correlates the sounds with Norma daintily proffering him her right hand, whereas Norma, on hearing the utterance, correlates it with her grasping his wrist with both hands and vigorously waving it in the air, successful communication would not be taking place. Moreover, this assumption of determinacy has bene®ted linguists in giving them grounds for calling linguistics a science; if there were no determinate entities then, the argument goes, there would be no way of systematically analysing the units which make up the language. Furthermore, if the units were not determinate, then one would feel compelled to take issue with Saussure and his successors' point of view that writing is a faithful representation of linguistic signs: `a language and its written form constitute two separate systems of signs. The sole reason for the existence of the latter is to represent the former' (Saussure 1922: 24±5). As these three points form the basis of twentieth-century linguistic theory, one sees again why the question of linguistic units is such a key one. Since a language manifests so many and varied viewpoints from which to identify the correlational patterns5, the main problem which confronts Saussurean theory is that of identifying such constants required for successful communication. The most intuitively plausible unit ± at least in western society ± is the word. Consequently, attempts to make sense of the concept `word' have helped determine the whole course of 4 This term is taken from Harris (1981) and refers to the view that communication is a matter of implementing one's knowledge of a language, which consists of no less and no more than knowledge of words which are strung together by a set of determinate rules. 5 e.g. does `red-haired' consist of one (red-haired), two (red+haired) or three (red+hair+ed) units? And if more than one, what is the meaning correlated with the `red'? Is it the same as the red in `I saw red'?

5

Orientation: The Word of Linguistic Theory

modern taxonomic6 linguistics, which has led to a plethora of terminological and theoretical muddles. The fate of the concept `word' in generative approaches is symptomatic of something more widespread in linguistics: the Saussurean attempt to develop explicit de®nitions which correspond to common-sense notions (intuitions of lay people), but at the same time are `scienti®c' in the sense of being adequate for purposes of rigorous descriptive analysis. It was Saussure who ®rst raised as a theoretical issue the status of the word7. Often termed the father of modern linguistics, he ®rst posed as an explicit problem the question of determining which (if any) of the linguistic units recognised in traditional grammar are units which have psychological reality, i.e. are units which `exist' in the mind of individual language-users, and he argued that only those units and distinctions meeting this criterion are valid for the purposes of a science of language. However, within his speech circuit model of communication every sign has to have a determinate form and meaning. There is a problem in characterising the variability of meanings as mentioned above, but also in locating various forms. Consider the word `to', an extremely common word, in the following three phrases: (a) I want to work, (b) I went to the job centre, and (c) Nine to ®ve. All speakers of the `same' language under the Saussurean speech circuit model would have at least to agree as to whether the instances of `to' are the same or different. If speaker A thinks that (a) and (b) are the same, whereas speaker B thinks that (a) and (c) are the same, then they are obviously not speaking the `same' language. Identifying ®xed forms in the spoken utterance is no easier, as in the problem of trying to locate `is' in utterances of the following types: [õnõt] (isn't it), [eõnt] (aint), [snát fE@(r)] (it's not fair) (Davis 1990: 8). 6 Taxonomic, as opposed to transformational generative linguistics, has been exclusively concerned with the classi®cation and description of languages by means of the segmentation of linguistic units, as in Bloom®eld's Language (1935). 7 Words had been discussed by earlier scholars (e.g. John Locke, G. Leibniz, Horne Tooke, Wilhelm von Humboldt) but they were not considered as autonomous `language' units: rather their introduction was to raise broader questions; for instance the origins of language, thought, knowledge etc.

6

Orientation: The Word of Linguistic Theory

The Word In addition to these problems outlined above, Saussure's position on the word was somewhat problematic and paradoxical. Although he claimed that the word, unlike the linguistic sign8, was not a psychologically real unit nor did it correspond to `our notion of a concrete unit' (Saussure 1922: 147), he yet went on to state that, despite the problems involved in its de®nition, the word is an intuitively recognisable unit (ibid.: 154). The difference, if any, between a psychologically real unit and an intuitively recognisable unit was nowhere discussed with any degree of clari®cation. Notwithstanding this problem, Saussure's arguments are doubly defective in that just because educated members of Western society can recognise and talk about words in a stretch of writing, that does not suf®ce to make words `real objects' in the sense of units of la langue9. Since Saussure, the status and de®nition of the `word' has become a controversial issue in twentieth-century linguistic theory. And `what is a word?' remains a perennially awkward question in contemporary linguistics, with no entirely satisfactory answer readily available (Crystal 1991: 329). This is a particularly problematic situation for some linguistic theorists who claim that understanding and producing sentences is a compositional process. the process by which a speaker interprets each of the in®nitely many sentences is a compositional process in which the meaning of any syntactically compound constituent of a sentence is obtained as a function of the meaning of the parts of the constituent . . . 8 The word, according to Saussure, was often a variable unit whereas a linguistic sign was a ®xed, determinate entity: `do', `does' and `did' are different forms of the `same' word having different forms and meanings, whereas they are three separate linguistic signs. 9 La langue, the object of the linguist's investigation, is the language structure, a system unaffected by the variations in its material manifestations, as opposed to langage, language in its entirety, which encompasses parole, individual speech.

7

Orientation: The Word of Linguistic Theory

. . . the syntactic analysis of constituents stops at the level of words . . . these are therefore the atoms of the syntactic system. (Katz 1966: 152±3) Linguists, faced with such insurmountable dif®culties, have tended to adopt one of three general attitudes towards the problem. One, a `part of the job' attitude, is to shrug it off as a kind of occupational hazard which linguists have to put up with: In linguistics there are a lot of such concepts which resist correct de®nition and, as a matter of fact, it can be said that linguistics is continually grappling with de®nitions. We have only to note how many de®nitions there are of the phoneme, syllable, sentence, etc. (Kramsky 1969: 8). A second, the `no big deal' approach, is to make light of the de®nitional dif®culty on the grounds that such problems do not really matter because, in practice, both linguists and ordinary speakers intuitively recognise words and therefore have no problems in identifying them. A third, the `distorted vision' approach, is to claim that many of our linguistic concepts (syllable, sentence, dialect) have fuzzy edges and, as long as there remain clear paradigm cases, the fuzziness is not a problem (ibid. 101). Thus: linguistically, and hence conceptually, the things in sharpest focus are the things that are public enough to be talked about publicly, common and conspicuous enough to be talked of often, and near enough to sense to be quickly identi®ed and learned by name; it is to these that words apply ®rst and foremost (Quine 1960: 1). And: The problem of the word must be viewed, above all, in terms of typical words, though we shall not avoid marginal problems. It is just these marginal problems that make an unambiguous de®nition of the word so dif®cult (Kramsky 1969: 9). 8

Orientation: The Word of Linguistic Theory

These attitudes both cast doubt on the notion of intuitive recognition and reject the Saussurean emphasis on acknowledging the language-user's own grasp of linguistic concepts (ibid.: 102). Such attitudes, in some cases, have led to theorists trying to convince us that words are really a sort of illusion ± some kind of sophisticated trick duping us most of the time. A striking fact about all speech is that all speakers ± no matter in what language ± are sure that they produce words, and all hearers are certain that they perceive utterances as sequences of words. When we pay attention to our own speech, we observe at once that we do not normally break up our utterances into words; rather, we run words together without intervening pauses . . . Does this mean that, because we do not pronounce utterances word by word, our perception that utterances are made up of words is a kind of illusion ± that we perceive words even though they are not actually there? I would argue that this is indeed the case, for what we hear is only partly determined by the physical signal that strikes our ears (Halle 1994: 335). Even though the anthropological linguist Edward Sapir did acknowledge the language user's individual grasp of linguistic concepts, (`there is not, as a rule, the slightest dif®culty in bringing the word to consciousness as a psychological reality' (Sapir 1921: 34)), he did not follow up this forthright claim by any further analysis of the ability in question or its cultural basis. The same is true for Sapir's and Saussure's successors. Many linguists wish to claim that the word is a more fundamental unit than you would suppose from the written word, and that the worddivision in written languages is not merely a convention. A case in point can be seen in the book Introducing Linguistic Morphology the purpose of which is to try to tackle the problem of word-forms in languages that have not yet adopted a writing system (Bauer 1988). 9

Orientation: The Word of Linguistic Theory

Speakers of English have, on the whole, a fairly good intuition about what `a word' is, although there are a few marginal cases where intuitions are unclear: . . . it could be pointed out that even preliterate children in our society have a reasonable notion of where words begin and end . . . There is also a certain amount of evidence . . . that non-literate speakers of unwritten languages know where words begin and end in their languages (ibid.: 45). But such views display an ignorance of the controversy about words as linguistic units. Over 50 years ago, Passy had produced some evidence that children and illiterates do not know where `words' begin and end in their languages citing such examples as `il marche a sait bien' instead of `il march assez bien' and `avecmane moisel' in place of `avec mademoiselle' (Passy 1932: 42)10. Despite such controversies, many chapters of linguistic textbooks have been devoted to de®ning the word, in particular attempting to develop a de®nition that includes not only nouns, verbs and pronouns but also the problematic de®nite article (Bauer 1988: 49)11. The problems inherent in delimiting a linguistic unit that, according to lay usage, apparently includes forms such as politician, liar, you, the and an but excludes forms such as ing, pre- and born loser, are intractable. But instead of a wholesale abandonment of the notion that the word is a basic unit of linguistic structure, linguists have either said that the problematic cases occur so infrequently that they should be regarded as `trivial' exceptions to the `rule', or declared forms such as the not to be `real' words (Lyons 1968: 204). A third, although less frequent, approach to these problems is to view the word as a theoretical construct: the `so-called concepts `the word' and `the sentence' have only systematic and empirical 10 My own children of 5 and 4, unversed in metalinguistic knowledge, con®rm Passy's evidence today. For example, ZoeÈ aged 4.7 months thought that icecream van was a word and Joshua aged 5.9 considered the group The Spice Girls to be a word. 11 In 1980, Roy Harris gave a critique of the standard tests of potential pausing, indivisibility and minimal free form for word criteria as rehearsed in all versions of Crystal's Dictionary of Linguistics and Phonetics

10

Orientation: The Word of Linguistic Theory

import with respect to a particular theory' (Botha 1968: 84). If the description and de®nition of linguistic units were only dependent on a particular theory for a particular (convenient) purpose, then it would not really matter how such units were understood or described. But unfortunately this is not the case: many linguists have higher aspirations: namely to de®ne a determinate unit which corresponds to native speaker intuitions. These aspirations are worrying in the respect that their postulated system of units cannot satisfactorily accommodate the English de®nite article and the French personal pronoun, both of which are well-established and common forms (Davis 1992: 105). Thus, although the word is a necessary component within the orthodox model of language and communication, its de®nition, despite much effort, has not been forthcoming. This is unfortunate for those scholars who wish to claim that linguistics is a science. Words in the Brain Other linguists, of the `psycho'- variety, have chosen to ignore the de®nitional problems, and have gone so far as to imply that this linguistic unit has a biological basis: all human languages have words, or wordlike segments, and there are general methods for compiling them in dictionaries . . . That such similar forms of communication could arise independently in every human group, however isolated, suggests that there is a biological basis for the human capacity for language (Miller, G.A. 1981: 4). Their assumption is that words `exist' in the brain as a `mental lexicon'. Such a lexicon is needed most immediately to produce and understand sentences, so lexical information must be represented in a manner 11

Orientation: The Word of Linguistic Theory

compatible with what people know about the syntax and semantics of sentences in their language (Miller 1978: 61). But this change of direction only leads to circularity (Davis 1992). The plausibility of the mental lexicon derives its support from the speech circuit model of communication, with additional `support' from the work on `speech errors' and `slips of the tongue'. This mental lexicon is supposedly a universal component of grammar since, as Aitchison claims, all languages divide words up into `parts of speech' or word classes (Aitchison 1987: 98). Fromkin shares this view, claiming that an explanation of the universal principles which are the basis of all languages is unobtainable without a modular12 conception of the mind. This conception sees knowledge of language as a system of independent modules or units each dealing with their own specialised functions: early performance models constructed to account for speech error data . . . depended on linguistic constructs. How could one begin to account for errors in features, segments, morphology, syntax, lexicon or semantics without reference to linguistic units and rules? . . . linguistic theory has guided the work of psycholinguists in their attempts to construct processing models with some reference to reality (Fromkin 1987: 3). The circularity of this assumption is glaringly obvious ± the evidence for postulating a mental lexicon in the ®rst place was supported by reference to speech errors etc. (Aitchison 1987), but here it is the speech errors themselves which are explained by reference to a mental lexicon. More worrying is this subtle change of means of explanation, without proffering any 12 A modular approach can also be found in such contemporary schools of linguistics as pragmatics. `The assumption that there is a distinction between linguistic and non-linguistic knowledge marks our approach as modular, and thus as consistent with the view of language found in Chomskyan generative grammar' (Blakemore 1992: 40).

12

Orientation: The Word of Linguistic Theory

justi®cation for such a modi®cation (Davis 1992: 106). Thus the de®nitional problem can, despite being recognised, be ignored. However, it was not cognitivists13 who ®rst made claims for the universality of the word, but, as mentioned earlier, the relativist and descriptivist Edward Sapir. Sapir's evidence, despite its apparent empirical foundation, is rather unconvincing: the naive Indian, quite unaccustomed to the concept of the written word, has nevertheless no serious dif®culty in dictating a text to a linguistic student word by word; he tends, of course to run his words together as in actual speech, but if he is called to a halt and is made to understand what is desired, he can readily isolate the words as such, repeating them as units (Sapir 1921: 34). Twice I have taught intelligent young Indians to write their own languages according to the phonetic system which I employ. They were taught merely how to render accurately the sounds as such. Both had some dif®culty in learning to break up a word into its constituent sounds, but none whatsoever in determining the words . . . I have obtained from one of these young Indians the words . . . isolated precisely as I or any other student would have isolated them (ibid.: 35, n6). The `evidence' in the ®rst paragraph is fundamentally ¯awed, in that it is unclear that someone without any form of alphabetic writing could yet have the concept of dictating a text word by word. All that is demonstrated here, as in the second paragraph, is that it is possible to elicit a desired response from an intelligent person when certain criteria are made clear. This in no way proves that the word is a psychological universal, however loosely this term is interpreted. 13 i.e. those linguists who believe that the abstract principles pertaining to the grammatical structure of all languages are encoded in the brain (e.g. post-1957 Chomsky).

13

Orientation: The Word of Linguistic Theory

For Sapir, psychological reality was simply opposed to physical reality, most clearly expressed in his analysis of the phoneme. In `Sound Patterns in Language' his aim was to show that `sounds and sound processes of speech cannot be properly understood in . . . simply mechanical terms' (Sapir 1949a: 33). He demonstrates this point with an analysis of the sound whmade in the context of blowing out a candle, and in a word such as when. Although the two sounds may be almost acoustically identical, they are not perceived to be so by speakers of a language, he claims, and therefore have no `psychological identity' (ibid.). Conversely, `not one speaker out of a hundred has the remotest idea that the t of a word like sting is not at all the same sound as the t of teem' (Sapir 1921: 44). And so, for Sapir, it is the direct judgements of a speaker which have `psychological reality': If the phonemic attitude is more basic, psychologically speaking, than the more strictly phonetic one, it should be possible to detect it in the unguarded speech judgements of naive speakers who have a complete control of their language in a practical sense but no rationalized or consciously systematic knowledge of it (Sapir 1949b: 47). With the advent of transformational generative grammar (Chomsky 1957), there was a rede®nition of psychology and a subsequent split between linguists and psychologists. It was felt that those psychologists concerned with language were able to explain only the learning process. `But there was nothing in learning theory which could account for the remarkable differentiation of verbal behaviour that transformational treatments revealed' (Fodor, Bever and Garrett 1974: 222). In endeavouring to validate the postulated transformational rules, linguists soon discovered that `direct judgements' of the speaker led to con¯icting results: and so, in the heyday of transformationalism, linguists were vying with each other to (re) write grammars that were, in some sense, `psychologically real'. This `some sense' required a rede®nition of the term `psychological reality'. 14

Orientation: The Word of Linguistic Theory

Chomsky's rede®nition of psychological reality was in part a defence of his early position (1965), that linguists should ®rst concentrate on devising a theory of linguistic knowledge rather than usage, thereby ensuring that the `messy' vagaries of talk were not allowed to infect the determinate stasis of knowledge. Chomsky claimed that the best available, currently undisproved (unfalsi®ed) theory is `psychologically real', or `correct', a position familiar to many natural scientists. For Chomsky, there is no distinction between `truth' and `psychological reality' (Chomsky 1980: 107). He considers linguists and natural scientists perfectly justi®ed in working with idealised systems. If a scientist wishes to discover, for example, why the sun's light gets converted to heat in investigating the nature of thermonuclear reactions that take place deep within the sun, it would be perfectly reasonable to come up with a theory based on the light emitted from the outermost part of the sun in constructing an hypothesis: it is senseless to ask for some other kind of justi®cation for attributing physical reality to the constructions of the theory, apart from consideration of their adequacy in explaining the evidence and their conformity to the body of natural science as currently understood (Chomsky 1980: 190). So in referring to a mental lexicon, Chomsky is invoking not a physical entity, but an idealisation: I . . . see no reason not to take our theories tentatively to be true at the level of description at which we are working, then proceeding to re®ne and evaluate them and to relate them to other levels of description, hoping ultimately to ®nd neural and biochemical systems with the properties expressed in these theories (Chomsky 1980: 107). We may ask whether the linguist's constructions are correct or whether they should be modi®ed or replaced. But there are few meaningful questions about the `reality' of these constructions ± their `psychological reality,' to use the 15

Orientation: The Word of Linguistic Theory

common but highly misleading term ± just as there are few meaningful questions about the `physical reality' of the chemist's constructions, though it is always possible to question their accuracy. At every stage of inquiry we try to construct theories that enable us to gain insight into the nature of the world, focusing our attention on those phenomena of the world that provide enlightening evidence for these theoretical endeavours. In the study of language we proceed abstractly, at the level of mind, and we also hope to be able to gain understanding of how the entities constructed at this abstract level and their properties and the principles that govern them can be accounted for in terms of properties of the brain. If the brain sciences succeed in discovering these properties of the brain, we will not cease to discuss language in terms of words and sentences, nouns and verbs, and other abstract concepts of linguistics, just as the chemist today does not refrain from speaking about valence, elements, benzene rings and the like. These may well remain the appropriate concepts for explanation and prediction, now forti®ed by an understanding of their relation to more fundamental physical entities ± or further inquiry may show that they should be replaced by other abstract conceptions, better suited to the task of explanation and prediction (Chomsky 1988: 7±8). The acquisition of language by children, it is claimed, also supports the notion of an idealised mental lexicon. Bauer, maintaining that speakers of English have a fairly good intuition about what a word is, states that this holds true even for preliterate children. He feels that they know where words begin and end because of their ability to ask questions of the sort `What does X mean?', where `X' represents what adults would term a word (Bauer 1988: 45). This evidence for a child's intuitive knowledge of `wordhood' is hardly conclusive. It merely shows that children have the ability to repeat an unfamiliar sound or sequence of sounds spoken by an adult. And even if a child in this sense could articulate every `word' in the Oxford English 16

Orientation: The Word of Linguistic Theory

Dictionary, that still would not prove that a child knew what a word was. The claim would just be a repetition of Sapir's error, i.e. an attribution of one's own concept to someone not in a position to grasp it. As Bowey and Tunmer write: `a child might be able to segment a speech sequence into words, thereby demonstrating an awareness of the `word unit' concept, without knowing what is meant by the term word. A child might also use the term word without sharing the adult's notion of what this term means' (Bowey and Tunmer 1984: 74). However, following Sapir, they still claim that the word has psychological reality: the ability to segment speech into word units does not appear to be dependent upon the prior ability to read [. . .] Words may thus be viewed as `psychologically real' units of language, independent of literacy. Nevertheless, the dif®culties involved in de®ning the word linguistically suggest that word awareness may not be easily acquired (ibid.: 73). Here it is unclear how one can ever know whether a child is using the term word in the same way as an adult, and consequently how one can ever be in the position to assign such a unit `psychological reality' however broadly that label is understood. Even if one accepts Aitchison's premiss, that one can speak of a mental lexicon only when a child develops the ability to classify objects ± somewhere between the ages of one and two ± (Aitchison 1987: 96), this still is no proof that a child knows what a word is. Piaget, despite his methodological shortcomings, shows that it is not until a child is about ten years old or so that his or her metalinguistic competence approaches that of an adult's. Piaget notes that in the ®rst two `stages' of development, the child confuses the sign with the thing signi®ed. Thus when asked whether any, and if so which, words were `strong', a child at the ®rst stage (six years or so) was unable to see the problem in the question, and replied `Daddy, because he's a daddy and he's strong'. Other `strong' words were `cloud' `because it gives light at night', and `umbrella' `because someone might poke you in 17

Orientation: The Word of Linguistic Theory

your eyes and that would kill you' (Piaget 1929: 57). Similarly `sun' was not considered to be `strong' `because it doesn't weigh much'. Piaget concludes that this experiment, by itself, does not prove very much, for there still remains the possibility that a child could realise what a word is without having the means of expressing the idea: `the word `word' implies for them the presence of the thing itself, in which case the experiment is of no value' (ibid.). He suggests that the only way of proving that the word is confused with the thing is to show that older children manage to understand the problem, although they are unable to solve it. An example of this is a child's reply to the same question, at a later stage of development: `Is a word strong?' `No, it can't do anything at all' `Are any words strong?' `Some words are strong' `Which?' `The word `strong' because you are saying it's strong' (ibid.: 58) This confusion between words and what they represent is offered as an explanation as to why children often reject function words or abstract words as words, `there being no concrete image to relate them to' (de Goes and Martlew 1983: 220). However, it is not just cognitive development which is lacking in children's lack of metalinguistic awareness, but rather, and more importantly, lack of exposure to and training in the use of the terms concerned (see Chapters 3±6 below). It is implausible to suppose that the various problems outlined above are due to the individual incompetence of the various theorists in question, or that there exists some de®nition of the word which so far no one has thought of, but which would, if hit upon, resolve all such contentious issues. The conclusions that one is forced to are, on the contrary, (a) that no satisfactory solution is available, (b) that, in spite of this, linguistic theorists from Saussure onwards have refused (for reasons not altogether clear) to discard the `word' and admit that it is simply not viable 18

Orientation: The Word of Linguistic Theory

as a unit for purposes of the `scienti®c' analysis of language, and (c) that much effort has been expended on trying to achieve a compromise between the demands of a rigorous objective de®nition and the observable facts concerning the everyday usage of the term14.

14 Even Halliday's social-functional approach is not immune to these criticisms above. Halliday's insistence that languages are existing systems which, in part, determine the linguistic options available, still depends on a prior notion of linguistic determinism. For a fuller critique of Halliday see Harris 1996 and Toolan 1996.

19

2 Methodology: The Word of the Layperson

As shown in chapter one, linguistic theorists have failed to come up with a single, internally consistent set of criteria which delimits all and only those items commonly designated in English as `words'. This failure has to do partly with the view that words are determinate with respect to both form and meaning. This view, in turn, is both underpinned by and underpins the assumption that communication is essentially a private matter of encoding and decoding the same set of speech signals tied to the same set of meanings: Language enables a speaker to transform con®gurations of ideas into trans®gurations of sounds, and it enables a listener within his own mind to transform these sounds back into a reasonable facsimile of the ideas with which the speaker began (Chafe 1970: 15). This Saussurean model of communication (code theory), as Taylor shows, is derived from attempting to place common-sense views of language (practical metadiscourse) on a scienti®c footing. it is precisely because of its mundane appearance that code theory is such a powerful form of intellectual discourse. It implies, that what the layman has been saying all along does in fact have a technical (or `scienti®c' in the lay sense of that term) justi®cation . . . code theory shows that . . . [our everyday metadiscourse] corresponds to the way things really are: that is to the truth (Taylor 1992: 71). 21

Methodology: The Word of the Layperson

But these so-called `common-sense' views do not conform to the `way things are'. In fact, if we suppose that communication is essentially a matter of transporting thoughts from one person's mind to another's by means of verbal utterances, we are left with the impossible task of identifying a successful, or even unsuccessful, episode of communication. How do we know what the thoughts are behind the utterance `what are you doing tonight?'. Code theory assumes that there must be `some private unobservable criteria . . . which account for what is observably successful communication' (Taylor 1981a: 271). But if the criteria for communication were private, how on earth would we know if successful communication and understanding were possible? But we often do judge when communication and understanding have been successful, and if there is some sort of communicational breakdown we can often identify the cause of the problem. In neither case do we need to appeal to the states of mind of the interlocutors, mental images or tacit linguistic knowledge. To claim that it is possible, in certain contexts, either to identify the cause of communicational breakdowns or to recognise a successful episode of communicational interaction does not underwrite the thesis that linguistic signs are determinate, ®xed etc.; such a claim can be reconciled with the `radical indeterminacy' of what is said and understood, despite the fact that some objectors appear to have common sense on their side. Trevor Pateman for example, raises some objections to Roy Harris's indeterminacy thesis that every linguistic act is integrated into the individual's experience as a new event which has never occurred before and can never occur again . . . The claim is not that speakers cannot produce or recognise instantiations of the same expressions on different occasions, but rather that this ability does not yield a criterion of demarcation between the linguistic and the non-linguistic, nor imply that whatever we say is decontextualizable. Repetition . . . is only partial replication, and even that partial replication is contextbound by succession in time. In what respects one utterance is a replication of another cannot be assessed independently of their sequentiality (Harris 1981: 55). 22

Methodology: The Word of the Layperson

Pateman objects on the grounds that to actually say that there is no repetition of any word cat, love, word etc., one has to use `overtly or covertly, and not merely mention a type, category or form cat, love, word' (Pateman 1987: 3). Although he speculates, we may be in the dark about how speakers and hearers produce and understand utterances . . . that they make use of some iterable form not de®ned by temporal co-ordinates seems presupposed by the very critique of that idea. In other words, forms like those `abstracted' by modern linguistics (Saussurean, Bloom®eldian, Chomskyan) are not necessarily the misleading and obscuring abstractions they appear to be to the empiricist [G. Sampson] and sceptic [R. Harris], but in some form must be psychologically real (ibid.). But Harris's point is not that speakers cannot recognise any distinction between types and tokens, nor that they may frequently entertain the possibility of saying `the same word', but rather that this ability, which is a highly sophisticated one facilitated (if not actually initiated) by the availability of writing, printing and standardisation, does not yield a `criterion of demarcation between the linguistic and non-linguistic'. Although types and tokens are, or may be, `real' for speakers, it is the speakers themselves who decide on any particular occasion on the identity of the types and tokens (see Hutton 1990, Love 1998). Thus the metalinguistic game of abstracting an utterance to subject it to any form of analysis (e.g. love is a four-letter word) is a highly sophisticated one relying on an arbitrary division between verbal and analytical description. For example, one could make, with the utterance `what are you doing tonight?', a number of varied analyses: `He asked me what I was doing tonight', `he asked me for a date', `he made a pass at me' or `he made a nonchalant comment about my activities this evening using non-descript, bland words'. In fact, lay speakers frequently do ask metalinguistic questions (`What does metalinguistic mean?') and make metalinguistic comments (`tsetse isn't an English word'). And herein lies the difference. It is the lay 23

Methodology: The Word of the Layperson

speakers who are making reasonable, purposeful comments and asking the sensible questions here; the linguists, however, are theorising about material (language) which started off as, and remains, an individual act of interpretation. It is therefore only when such metalinguistic games are taken as constants for the purposes of theorising about a language that they constitute the misleading and erroneous activity condemned by Harris. But this is not to deny that such a mistaken, ®xed-code model of communication has an air of plausibility which stems in part from our metalinguistic talk subsumed by the `conduit metaphor' (Reddy 1979). Although phrases such as `Try to get your thoughts across better', `Don't force your meanings into the wrong words' and `your words are hollow- you don't mean them' (ibid.: 286±288) seem innocuous enough, Reddy identi®es them as adversely affecting `thought processes'. The major framework of this `conduit metaphor' has four components: (1) language functions like a conduit, transferring thoughts bodily from one person to another; (2) in writing and speaking, people insert their thoughts or feelings into the words; (3) words accomplish the transfer by containing the thoughts or feelings and conveying them to others; (4) in listening or reading, people extract the thoughts and feelings once again from the words (Reddy 1979: 290). which, by objectifying meaning `in a misleading and dehumanising fashion', in¯uencing us `to talk and think about thoughts as if they had an external, physical reality, trivialises the role of the reader/listener' (ibid.: 308). This metaphor, in turn, is underpinned by certain communicational and cultural practices, such as those of legal procedure, which often takes the form of a debate over the meaning of certain words, e.g. whether or not President Clinton was guilty of perjury, whether or not exhibit Y is a weapon or piece of cutlery. The fact that the dictionary is often the ®nal `court of appeal' regarding our linguistic practices, whether it be for deciding on a game of Scrabble, for checking 24

Methodology: The Word of the Layperson

the spelling in an examination script, ®nding an equivalent word in French or deciding whether or not the president of the United States committed perjury when he claimed `I never had sexual relations with that woman' inculcates, or certainly reinforces, the view that words have ®xed forms and ®xed meanings. Obviously, the dictionary alone is unable to settle legal disputes. This can clearly be seen in the report by Independent Counsel Kenneth Starr on the grounds for President Clinton's impeachment1. Although Monica Lewinsky performed oral sex on President Clinton on more than one occasion, although Clinton fondled Lewinsky's bare breasts and genitalia with his hands, although Clinton inserted a cigar into Lewinski's vagina, although they both engaged in oral-anal contact and had phone sex, and although Clinton frequently ejaculated, Clinton still denied that he had a sexual relation with Lewinsky. How he was able give a sworn testimony and deny that is interesting for the purposes of this discussion. In the ®rst place, Clinton, although he had recourse to a dictionary, decided to opt for his own de®nition of `sexual affair' and `sexual relationship'. For Clinton's purposes he said that both acts require `sexual intercourse, no matter how extensive the sexual activities might otherwise be'. And also, with `sexual relations' that too requires sexual intercourse. `According to President Clinton, oral sex does not constitute sexual relations'. How he managed to arrive at this argument is revealing in that it shows up the inadequacies of lexicographical practice and the concomitant credulity of the readers of such `authoritative' texts. In Clinton's civil deposition, `sexual relations' was de®ned as `when the person knowingly engages in or causes contact with the genitalia, breasts, or anus of `any person' with an intent to arouse or gratify the sexual desire of `any person'. Concerning oral sex, the President's sole answer to the charge that he lied under oath at the deposition focused on his interpretation of `any person' in the de®nition. Ms 1 The report cited here is from its electronic form. These ®les may be retrieved at http://www.royalriver.net/report/

25

Methodology: The Word of the Layperson

Lewinsky testi®ed that she performed oral sex on the President on nine occasions. The president said that by receiving oral sex, he would not `engage in' or `cause' contact with the genitalia, anus, groin, breast, inner thigh, or buttocks of `any person' because `any person' really means `any other person'. The president further testi®ed before the grand jury: `[I]f the deponent is the person who has oral sex performed on him, then the contact is with ± not with anything on that list, but with the lips of another person'. (p. 10 Starr report) [my emphasis]. Thus `sexual relations' or `oral sex' is not something that can be decided by a purely linguistic analysis of the phrases. Similarly, no phonetician can tell a poet whether blasphemy rhymes with gas leak, and nobody has been able to settle, to everyone's satisfaction, the dispute about whether The Satanic Verses is a blasphemous work. The fact that there are so many languagegames2 to be played with ostensibly the `same' word, should indicate the necessity for a radical break with the idea that language always functions in one way ± that of encoding and decoding thoughts into linguistic forms, as though any pragmatic consequences resulting from this mental activity were extraneous to the linguistic act and not part of language. Nor are the terms word, language, sentence etc. different in kind from the words oral sex, blasphemy weapon or murder. The terms word, sentence, meaning are just ordinary words having ordinary, mundane uses. In his later writings, Wittgenstein was reacting against his former work, the Tractatus Logico-Philosophicus and also those philosophers who were misled into asking meaningless questions by the very terms available in a language. When I talk about language (words, sentences, etc.) I must speak the language of everyday. Is this language somehow 2 The concept of `language game' was extensively developed in Wittgenstein's Philosophical Grammar (Basil Blackwell 1969) showing the connection between the speaking of language and non-linguistic activities.

26

Methodology: The Word of the Layperson

too coarse and material for what we want to say? Then how is another one to be constructed? ± And how strange that we should be able to do anything at all with the one we have! (Wittgenstein 1974: 77). Instead of seeking the common element of all that we call `language', `word' etc., Wittgenstein claimed that such metalinguistic concepts do not necessarily have one thing in common which makes us use the same word for all instances. Instead, the similarities between instances covered by such terms are akin to the similarities between members of a family, and he therefore characterised the similarities as `family resemblances' (Wittgenstein 1958: 67). Rejecting the idea that language always functions in the same way, Wittgenstein also proposed the conception of language-games, bringing `into prominence the fact that the speaking of language is part of an activity, or of a form of life' (ibid.: 23). An appropriate way of investigating the ordinary use of language is to examine its use by ordinary, i.e. not `ideal' speakers, by your local ®shmonger, MP, house-husband or college drop-out. Such an approach, instead of starting with an analysis or dissection of language, takes as its aim what these ordinary speakers understand by what is going on linguistically. Such an understanding will depend in important respects on the speci®c cultural contexts into which they are integrated. What is just as important as culture-speci®c aspects of what is understood as language is the integration of cultural artefacts with those aspects of communication which are universal to humanity ± i.e. whatever happens when at least two people interact with mutual awareness. Silence can show understanding, just as much as it can show indifference or anger; and the fact that being silent for the purpose of communicating is a voluntary act also shows that it is just as much a linguistic act as shutting the door when requested, with or without verbal utterance, deliberately ignoring someone when asked to shut the door, or telling someone to shut it herself: `If you exclude the element of intention from language, its whole function then collapses' (Wittgenstein 1975: 20), `I will count any fact whose 27

Methodology: The Word of the Layperson

obtaining is a presupposition of a proposition's making sense, as belonging to language' (ibid.: 45). But since in some cultures there are some communicational channels not available in other cultures, such as writing, the forms that integration will take will be partly cultural : `an intention is embedded in its situation, in human customs and institutions' (Wittgenstein 1958: 337). The implications of this for an analysis of the word are that since different individuals and the same individual will have and have had different experiences, and since they will be attempting to integrate such experiences to achieve a particular goal at a particular time in a particular context, there is no question of the `words' they use being invariant linguistic units. Any question as to what a `word' is will therefore presuppose certain cultural practices, but the answer will depend on the language-game into which the question is integrated. None of this is a denial that we can recognise `the same word'; rather this ability can now be explained by reference to our normative linguistic practices, and not by reference to some invariant units stored in some abstract mental lexicon. Stressing the `normativity' of our linguistic practices is meant to bring out not only the regularity of certain aspects of language (`plurals are usually made by the addition of an s') but also the justi®cation for and evaluation of using, for example, one linguistic feature over another (`criteria is the plural of criterion because it is a Greek word' or `that's what my English teacher told me' etc.). Thus this conception of language contrasts with the orthodox view of linguistic determinacy, autonomy and objectivity, thereby presenting a framework different from the conduit metaphor as identi®ed by Reddy above. In common with other social practices `it emphasises the location of the voluntary acts of individual linguistic agents within the coercive moral context of everyday life' (Taylor 1990: 138). Thus Through normativity we make language signi®cant; and, from the perspective of this signi®cance, language has form' (ibid.: 143 [my emphasis]) 28

Methodology: The Word of the Layperson

If the products of verbal interaction are amenable to formalisation, this is because the context-dependent activities of producing speech are normative activities. The patterning of language is of our own making: to explain it we must examine it in the making (ibid.: 147). Such a signi®cance-determining perspective makes issues like word-identi®cation open to empirical treatment through an analysis of the integration of the cultural in¯uences on an individual, in conjunction with what he or she is aiming to do in a given situation. Accordingly, the theoretical framework is an integrational one, based on Harris 1996, 1998. An Integrational Approach An integrationist approach is in opposition to a segregational approach, the latter having dominated the ®eld of twentiethcentury linguistics. Segregational models of signi®cation `focus on features and relationships which de®ne a sign irrespective of its contextualization' (Harris 1996: 145). What is in contention as between the two approaches is whether, or to what extent, linguistics is entitled to decontextualize human linguistic behaviour in order to isolate, describe and explain various aspects of it. For the integrationist, all decontextualization distorts, and therefore the resultant linguistic descriptions and explanations, to the extent that they rely on decontextualized `data' are automatically suspect. For they are no more than methodological artifacts of the oversimpli®cations from which they proceed (Harris 1998: 13). Accordingly, an integrationist does not see language as something divorced from other activities, whether verbal, graphic, gestural, pictorial . . . If that is tantamount to denying the existence of an object language so be it; for in reverse of the segregationist belief, languages presuppose communication: 29

Methodology: The Word of the Layperson

for the integrationist the study of communication comes ®rst, because unless we can ®rst analyse the relevant communicational process(es) we have no basis for constructing any rational account of what `a language' is (Harris 1998: 11). For the integrationist, the possibilities and limits of human communication both in general and in particular instances are governed by three kinds of factors: the biomechanical, the macrosocial and the circumstantial. Biomechanical factors relate to the physical and mental capacities of the human being. Macrosocial factors relate to practices established in the community or some group within the community. Circumstantial factors relate to the speci®cs of particular situations (Harris 1998: 29). And these three factors today, between them suf®ce to identity any human enterprise whatsoever. the fact that A and B communicate in speech only via sounds of a certain amplitude and frequency is a biomechanical factor, having to do with the physiological constitution of the human body. The fact that A and B cannot communicate in Swahili because B knows no Swahili (even though A does) is a macrosocial factor. The fact that A can speak to B even though separated by a distance of thousands of miles (because a telephone is available) is a circumstantial factor. It is possible to consider biomechanical, macrosocial and circumstantial factors independently of one another. But any episode of communication will involve the integration of all three. That is to say, communication in any form will impose on the participants requirements of a biomechanical, macrosocial and circumstantial nature and to organize their participation in such a way that these requirements do not con¯ict (Harris 1998: 30). 30

Methodology: The Word of the Layperson

An integrationist approach rejects the assumption of generativist linguists that linguistic units are to be explicated in terms of postulated rules and procedures lying far beyond the reach of consciousness. In so doing it holds no brief for the semantic concerns of philosophers such as Frege, Russell, Dummett and Davidson in their attempts to formulate a theory of `meaning'. Such philosophical enterprises, in attempting to construct a theory of meaning `for a natural language' (i.e. a theory of È bersicht (surview) of the various understanding), lack an U criteria by which we would customarily bring to bear in showing that we have understood another's request, command, statement etc. For an integrationist, `word'-recognition is word recognition, i.e. involves a speci®c metalinguistic term, its use being seen as a function of speci®c cultural practices institutionalised in a given cultural context. This study is an attempt to make an empirical study of the relevant cultural practices involved in wordidenti®cation and word-characterisation and to draw up theoretical conclusions concerning the status of the linguistic unit in question. Recognising that the ®xed-code model of communication bolsters and is bolstered by various communicational practices and ways of talking about language, this chapter will indicate various forms integration takes. The concentration is upon lay metalinguistics over technical metalinguistics, since from an integrationist perspective linguistics is seen as essentially lay-oriented. The methodology is accordingly broadly based on that used in the social sciences. Empirical Evidence and Methodology The recognition of linguistic units can be observed in the social practices of language-users and speci®cally in their metalinguistic practice, including the use of the term word and associated vocabulary. For an analysis of this particular unit, I asked direct questions of twenty-one informants all personally known to me. This type of `ask-the-speaker' approach is typically avoided in segregational linguistics. One common objection to soliciting lay judgements has been that they are, more often than not, 31

Methodology: The Word of the Layperson

erroneous, since there is often a contradiction between what native speakers say that they do and what is actually observed by the linguist. As interesting as these reactions may be for the study of `tertiary attitudes towards language', they intervene between us and the linguistic facts, since they are not the fundamental object which we set out to study (Labov 1975: 41). Furthermore, a lay native-speaker is not an `expert' in the area of linguistic judgements, and the responses elicited by direct questioning of subjects might re¯ect more the way they think things ought to be, rather than the way they are. The speech community member, after all, is being asked to do on-the-spot introspection without the bene®t of education in the methods of anthropological research (Fasold 1990: 49). And as Bloom®eld noted, in an early discussion of this issue in modern linguistic theory, a subject, if corrected by a linguist, is likely to produce an unfriendly tertiary response. The tertiary response occurs almost inevitably when the conventional secondary response is subjected to question. The tertiary response is hostile; the speaker grows contemptuous or angry. He will impatiently reaf®rm the secondary response, or, more often, he will resort to one of a few well-formed formulas of confutation (Bloom®eld 1944: 49). Bloom®eld also mentions this en passant in Language when he discusses whether the ®rst syllable of cranberry has a meaning: As a practical matter, observing language in the ®eld, we soon learn that it is unwise to try to elicit such forms; our questions confuse the speakers, and they may try to get rid 32

Methodology: The Word of the Layperson

of us by some false admission, such as, `oh, yes, I guess cran means red' (Bloom®eld 1935: 160). Some even adduce a psychological explanation as to why the metalinguistic judgements of lay people are not to be trusted; that cognitive processes involved in making metalinguistic comments are more complex than those processes involved in ®rst-order linguistic tasks: giving language judgements ± retrieving and making use of one's own intuitions ± is relatively hard, compared to talking and understanding. Thus it is not surprising that we ®nd extensive individual and population differences in performance on the harder judgmental tasks, compared to lesser differences in talking and understanding. We believe this is because judgmental performances require a higher order of self-consciousness than do speech performances. To give a language judgement, one must take a prior cognitive process (linguistic performance) as the object of a yet higher-order cognitive process (re¯ection about language performance or, as we have called it, metalinguistic performance) which may have properties of its own (Gleitman and Gleitman 1979: 105). Within linguistic description and theory, only two approaches have been considered valid: the linguist either describes the language `on the basis of objective facts' or explains `the language faculty through the study of intuitions' (Labov 1975: 6). An exclusive concentration on one or both of these approaches has caused a whole dimension of linguistic evidence to be omitted. As has been pointed out by Nigel Love, the omission of this dimension has farreaching theoretical implications: What requires consideration is the extent to which it makes sense to suppose that there can be an objective account of what language is, independent of what language-users demonstrate, by their way of using it, that they think it is. No attempt to ground an analytic understanding of how 33

Methodology: The Word of the Layperson

language works can afford to ignore the question of how far, in seeking such an understanding, the linguist can or should emancipate himself [sic] from preconceived views on that issue which may themselves actually be constitutive of certain kinds of communicational practice (Love 1998a: 66±67). This is not to deny that the possibility of a ®eld of inquiry concerned with lay metalinguistics had ever been contemplated. For instance, in 1959 Grootaers showed that by questioning native informants one could identify dialects purely subjectively. And in 1964 Hoenigswald wrote the necessity to warn against confusing the facts of speech with the talk about speech has been too real to allow much serious interest in the latter (Hoenigswald 1964: 17). If other phases of social science are any indication, we should be interested not only in (a) what goes on (language), but also in (b) how people react to what goes on . . . and in (c) what people say goes on . . . It will not do to dismiss these secondary and tertiary modes of conduct merely as sources of error (ibid.: 20). But the theoretical implications of this type of study for linguistic theory were never a matter for consideration; nor was the possibility of incorporating such ®elds of inquiry into linguistic theory in any systematic way ever seriously entertained. (Cf. Grootaers' statement: for the present-day speaker the opposition to the neighbouring area is mainly psychological and traditional; it is completely valueless as a starting point for the linguistic study of dialect units and dialect boundaries (Grootaers 1959: 384)). The situation in linguistics is thus quite unlike that in many areas of the social sciences in which the 34

Methodology: The Word of the Layperson

`practical theorizing' of laymen cannot merely be dismissed by the observer as an obstacle to the `scienti®c' understanding of human conduct, but is a vital element whereby that conduct is constituted or `made to happen' by social actors (Giddens 1976: 52±3). Barnes observes that `the philosophy of social science is too important a matter to be left to the philosophers' (Barnes 1990: 2), and that its subject-matter, unlike in the natural sciences, is `made up of phenomena that are familiar to the general public' (ibid.: 4). My study aligns itself with these viewpoints in the social sciences. Its investigation of the word assumes with the social scientist that `a social investigator draws upon the same sorts of resources as laymen do in making sense of the conduct which it is his [sic] aim to analyse or explain' (Giddens 1976: 52). Following McGregor 1982, 1983, 1984, 1985, 1986a, 1986b, 1990; McGregor and White 1990; Kreckel 1981; Gumperz and Tannen 1979; Gumperz 1982, the present investigation acknowledges social differences in language variation as recognised by traditional sociolinguistic inquiry, but (in part because of the concentration on lay judgements) marks a break with its traditional methodology. the correlation of linguistic variables with social variables has begun from the assumption that social groups are identi®able and known. This, however, is an issue much in dispute in the social sciences. That is, the question of what a social group or subgroup is has been very problematic, especially in urban areas, where much sociolinguistic work has been focused (Gumperz and Tannen 1979: 306). Gumperz and Tannen suggest a way of approaching the problem outlined above which (a) avoids a priori identi®cation of social groups, but rather builds on empirical evidence of conversational cooperation; and (b) extends the traditional linguistic 35

Methodology: The Word of the Layperson

method of in-depth hypothesis testing with key informants, to the process of conversational inference (ibid.). They adopted an ask-the-speaker approach eliciting their informants' own interpretations of communicational utterances. Social boundaries can then be empirically determined, as a result of the data gathered, based on a similarity of interpretations and agreement as to what aspects of the communication led to those interpretations (ibid.: 308). McGregor's work on what individuals have to say about `what is going on' in their own and other people's verbal exchanges, although in sympathy with Gumperz and Tannen's ask-thespeaker approach, is not committed to, and in fact rejects, their assumptions that listeners know what linguistic features led to the interpretation of a certain utterance (McGregor 1983, 1986b). Similar problems, he feels, beset Kreckel's work: `it would be quite wrong to assume that the kind of `messages' communicated in conversation are simply those which the analyst has sought to elicit' (McGregor 1983: 276). By emphasising the radical indeterminacy of what was said and understood, McGregor heralds a complete break with traditional sociolinguistic inquiry in particular and positivistic linguistic theory in general. In so doing he appears very much in sympathy with an integrationist approach. The advantages of directly asking the speaker about the term word are manifold: being essentially familiar to the layperson of Western culture, it is no different in kind from the concepts studied in sociological inquiry, and thus the methods of sociological, in particular ethnomethodological, analysis are just as applicable to this metalinguistic term: `it is reasonable to assume that if a linguistic phenomenon is spoken about by native speakers, it is worth investigating' (McGregor 1982: 125 and 1983: 276). There is no question of supposing that the word has a uniform interpretation for all speakers of English, and a fortiori no question of attempting to determine what that interpretation 36

Methodology: The Word of the Layperson

might be. By the same token, questions of a statistical order relating to the size and typicality of the corpus of data become irrelevant since, as in all current ethnomethodological research, the generalisations drawn from an analysis of the data do not depend for their validity on projections drawn from a sample. In principle, evidence from a single informant would suf®ce as adequate data. Although this is also a tenet of many segregational linguists, in their case it is based on the assumption that all speakers are carbon copies of one other, and thus the linguist can use his or her own intuitions (see Chomsky 1965: 18). This is also one of the `paradoxes' Labov detects in orthodox linguistics: the social aspect of language can be studied through the intuitions of any one individual, while the individual aspect can be studied only by sampling the behaviour of an entire population (Labov 1972: 105). For an integrationist, what matters is membership of the linguistic community, not typicality of membership (which is, in any case, a notion of extremely dubious theoretical validity (Durkheim 1895)). It follows that in the methodology of this study there is no question of establishing `objectively the social class of the informants' (Downes 1984: 88) and no tacit appeal to a `consensus principle' (Labov 1975). On the contrary, an integrational approach to the problems of identifying linguistic units pursues to their logical conclusion the implications of the Saussurean dictum that in linguistics `it is the viewpoint adopted which creates the object' (Saussure 1922: 23). Some may object that this approach would lead to a regress of interpretation, that a linguist providing his or her interpretations of informants' interpretations of the purpose etc. of an experiment, would be counter-productive, i.e. would not lead anywhere. But as McGregor points out, not only is such a `regress' necessarily part of any kind of linguistic analysis (empirical or not) but also, as Gumperz demonstrates, can lay bare `the communicative processes that underlie categorisation, intergroup stereotyping, evaluation of verbal performance and 37

Methodology: The Word of the Layperson

access to public resources in modern societies' (Gumperz 1982: vii). Such a regress, unlike some others, actually mirrors what language is and how it works. The Present Study and Questionnaire The method I employed for an analysis of the word, involved taped interviews with twenty-one informants. These interviews were personally conducted and transcribed over a year. (Informant pro®les are given at the end of this chapter). All the informants were personally known to me. This presumably eliminated inhibitions of the kind that many informants have when being taped by a stranger. The questionnaire (see below) was used as a guideline which could be elaborated upon, as informants would often spontaneously volunteer information on the topic. Some informants who knew each other well were recorded together, as the pilot study had indicated that productive discussions frequently developed in those circumstances. This technique was frequently used by McGregor in order to `minimise my in¯uence as analytic observer and thus encourage greater spontaneity of comment' (McGregor 1986b: 161). As in McGregor's studies, but unlike those of Gumperz and Tannen, informants were not always given speci®c instructions concerning what to comment on. Each question began with the underlined question on the questionnaire. The informants were told that they were welcome to make any comments at any time, and that they could expand on their answers as much as they wished. In some instances informants may have offered comments simply `to ful®l the informant role in which they have been placed' (McGregor 1986b: 158). But this hazard, if it is one, is intrinsic to asking questions. In everyday interaction, communicators are placed in similar question-answer situations ± market surveys, appraisals etc. ± where it is impossible to determine with certainty whether the question has provoked a response which would otherwise never have been consciously formulated, let alone articulated. 38

Methodology: The Word of the Layperson

Questionnaire Question Number One

How many words are in these sentences? 1) 2) 3) 4) 5) 6) 7) 8) 9) 10) 11)

Do you want to go windsur®ng? No, I don't I've got to go to London The Queen of England's palace is there John's got a new car I have a new washing-machine Have you got a colour TV? Let's play hide and seek 1,2,3, go! Do you know your ABC? No, it is imbloodypossible

This question was conducted in two parts: I ®rst read these sentences aloud, then later, when all the questions had been completed, the same question was asked when the subjects had the written sentences in front of them. The aim of this question was to investigate a lay person's criteria for detecting wordboundaries and to ascertain whether different criteria are employed in writing and in speech. Question Number Two

Are the underlined words the same words or different words? 1a) He cashed a cheque at the bank b) He slipped on the bank and fell in the river c) I work in a blood bank 2a) The word nothing is a noun b) Nothing is cheaper than Brand X washing-powder 39

Methodology: The Word of the Layperson

3a) My neighbour is a pervert b) Don't pervert the idea 4a) That man works in a bank b) Man is a human being 5a) A dog has four legs b) Dogs have four legs 6a) You may borrow my car b) The road may be blocked 7a) Anyone can make a mistake b) Can I smoke c) He can speak English 8a) He must be back by midnight b) He must have gone for a walk 9a) I walk to school b) I walked to school 10a) I break everything b) I broke everything 11a) I am a nurse b) I was a nurse The informants had the questionnaire in front of them throughout this question, which was to investigate how they perceived the identity of a word and how they justi®ed their perception(s) Question Number Three

How would you compare or contrast these pairs of words? It doesn't matter if you are not sure of the meaning. 1) antidisestablishmentarianism it 2) ouch hey 40

Methodology: The Word of the Layperson

3) hereinafter witnesseth 4) alas lest 5) epistemological ontogenetically 6) amen thee 7) blimey shit 8) cuckoo crisp 9) parsimonious stingy 10) tsetse phthisis 11) gazooks snortle 12) old mature The informants had the written sentences in front of them. This question was an investigation as to how lay speakers characterised words ± e.g. the epithets and adjectives applied to such pairs. The words were paired thus so as to give an idea of various dimensions that could be commented on (e.g. long/short, old/new) although the purpose was not necessarily to produce perceptions matching those of mine when I devised the questionnaire. Question Number Four

Comment on how the underlined words are used in these sentences 1) John is a pig 2) He aint a pig 41

Methodology: The Word of the Layperson

3) 4) 5) 6) 7) 8)

Are those glasses dead? He was mad at me We was only playing My husband and me My husband and I It's em a nice colour

This was an investigation into a lay informant's understanding of `grammaticality' and `metaphorical' quality. Question Number Five

Do you think that all words must have a meaning? If so, what about these words? To, the, so, it, number Question Number Six

What is the difference between the underlined words? 1) I have a watch Watch your language Sit on that chair I will chair the meeting 2) What's your fancy That's a fancy dress 3) I love dancing Hello love That's a lovely picture He looked at her lovingly This was to examine the informants' understanding of the `parts of speech' classi®cation and to examine what, if any, criteria were employed in the assignment of a particular word to a word class. 42

Methodology: The Word of the Layperson

Question Number Seven

Comment on the de®nitions of these words taken from `The Concise Oxford Dictionary' 1) CAT: `Small furry domesticated carniverous quadruped' 2) YOGA: `Hindu system of philosophic meditation and aestheticism designed to effect reunion with the universal spirit' 3) ONTOLOGY: `Branch of metaphysics dealing with the nature of being' 4) ANT: `Small social hymenopterous insect proverbial for industry'. This was an investigation into the informants' attitude to the dictionary, `meaning', and `de®nition'. Question Number Eight

Comment on the spelling of any of the words in these sentences 1) Lots of luv from John 2) What color is that? 3) a) Do you realise what I'm saying? b) Do you realize what I'm saying? 4) He's a shef at the Ritz 5) Right me a letter 6) I'm singin' in the rain 7) `Kool and the Gang' are a good pop group 8) I think queen elizabeth is digni®ed This, an investigation into `linguistic form', was to ascertain the lay person's views on orthographic practice. 43

Methodology: The Word of the Layperson

Question Number Nine

This question was presented orally. What does the word `cat' begin with? And the word `mat'? How do the letters that begin the words `cat' and `mat' differ from, say, the letters that begin the words `ate', `eat, and `it'? Does the word `cat' begin with the same consonant as the word `kill'? What consonants do these words begin with: `psychology', `cycle' and `sight'? How many consonants and vowels in these words? `Sheep', `complete', `dumb'? This was also an investigation into linguistic form, but in this case it was to examine how informants identi®ed the parts of words not normally thought of as meaning-carriers. Notes in the Transcriptions All the taped interviews were transcribed. My questions and utterances are designated by the numeral 1, the informants' by letters. In the transcriptions, punctuation is introduced only to make certain passages intelligible. Underlined words or phrases represent particular words, phrases or sentences under investigation. A stressed word or sentence is transcribed in capital letters. A series of dots signify editorial omission of irrelevant transcription and a short hyphen represents a pause in the conversation. Informant Pro®les Informant A British male age 35, a landlord of a public house. Educated to an HND in electronic engineering. He was formerly a Staff Sergeant in the army. Reads The Morning Advertiser and The Racing Post. 44

Methodology: The Word of the Layperson

Informant B British male age 32, production engineer with 6 O levels, 2 CSE's and 2 A levels. Reads popular `Top 10' paperback ®ction. Informant C British male age 24, a postgraduate in history (doctoral status) at Oxford University. Reads a wide range of `academic' and ®ctional books. Informant D Northern Irish male age 22. A postgraduate in classics (doctoral status) at Oxford University. Reads a wide range of `academic' and ®ctional books. He is a practising Catholic. Informant E British male age 27, unemployed and rarely attended school as a child. Only reads letters. Informant F Scots female age 30, sometimes works in a public house owned by her `common law' husband. Reads Mills and Boon-type ®ction. Informant G British male age 50, dialectologist, historical linguist and a Reader in English language at London University. Reads `academic' texts and ®ctional novels. Informant H American male age 23, a law student at an American University. Holds a BBA in ®nance and public administration. Reads `academic' texts and ®ctional novels. Informant K British male age 38, unemployed with no quali®cations. Does not read anything. Informant J British male age 65, works as a manager in a leather-goods shop. Left school at 14. Reads The Daily Mail and The Daily Express. 45

Methodology: The Word of the Layperson

Informant L British female age 36, part time cleaner with two O levels. Reads The Sun newspaper. Informant M British female age 19, a sales assistant. Has four O levels. Reads `Top 10' ®ctional paperbacks. Informant N British female age 21, a secretary. Has 7 O levels. Reads `Top 10' ®ctional paperbacks. Informant O British male age 35, no formal education. Works in a garage owned by his father. Reads The Sun newspaper. Informant P British female age 20, with 2 CSEs. Currently doing casual temporary work. Reads Cosmopolitan. Informant Q British male age 22, postgraduate in linguistics (doctoral status) at Oxford University. Reads a wide range of `academic' and biographical books. Informant R Zimbabwean male age 24, postgraduate studying Agriculture at Oxford University. Reads `academic' texts. Informant S British male age 25, ®nancial consultant who failed his BSc in Biochemistry at London University. Reads The Times and The Financial Times. Informant T British female age 20, circus entertainer who left school at 16 without any formal quali®cations. Reads tabloid newspapers. Informant W British male age 27, unemployed and rarely attended school as a child. Reads `male magazines'.

46

Methodology: The Word of the Layperson

Table 1 How many words are in these sentences? QUESTION NUMBER ONE Qu. Inf. A B C D E F G H J K L M N O P Q R S T V W

01 66 66 6± 76 66 66 66 66 66 6± 6± 76 ±6 66 76 66 66 66 66 66 76 LR

02 33 44 3± 33 33 33 33 33 33 3± 3± 34 33 22 33 33 33 33 33 33 33 LR

03 66 76 6± 66 66 66 66 66 66 6± 6± 76 76 65 66 67 66 66 66 66 66 LR

04 77 77 7± 77 77 77 77 77 77 7± 7± 7± 77 77 77 78 77 77 77 77 77 LR

05 55 65 5± 55 55 55 55 55 55 5± ± ± 65 55 44 55 56 55 55 55 55 55 LR

06 55 55 5± 65 54 66 55 66 66 7± 6± 65 66 44 66 55 55 65 55 66 66 LR

07 08 09 66 55 44 66 46 44 6± 3± 4± 76 55 44 64 55 11 66 55 44 66 33 44 66 55 44 65 55 44 7± 5± 4± 6/7 ±5 ± 4 ± 66 5± 41 66 55 44 54 55 11 66 55 41 66 36 44 66 55 44 65 55 41 6 6 3/5 5 4 4 66 55 44 76 55 44 LR LR LR

10 11 55 44 55 44 5± 5± 77 44 44 44 75 44 55 44 75 64 74 53 7/5 ± ± ± 7± 6± 7 7 5/7 7 55 54 44 43 44 55 55 44 55 54 54 64 4 4 5/3 3 75 44 75 64 LR LR

Key: The x axis, labelled Qu. 01, 02 etc. represents the question number. The y axis, labelled Inf. A, B etc. represents the informant identi®ed. The ®rst number in the grid represents the number of words the informants counted when listening to the question, the second when the informants read the sentences themselves. Two numbers noted for a question means that two responses were given. The lack of response on certain questions signals that the informant did not complete the questions (C: `I haven't really got anything to say on these de®nitions so I should stop'), in this instance through boredom.

47

Methodology: The Word of the Layperson

Table 2 Are the underlined words the same or different words? QUESTION NUMBER TWO Qu. Inf. A B C D E F G H J K L M N O P Q R S T V W

01 s ? bd d s d s bd d s s bd s d bd d d bd bd s bd

02 s s s s s d s d s s s s s s s s d d d s d

03 s d d d s s s d s s s s s d s d s s s d d

04 s s d d s d s d s d d d d s s d d d d d s

05 d s d d d d s d d s ? s s s s d d s d d d

06 s s d s s d s s d s s d d d s d s s d s bd

07 s bd d s s s s bd d s s ds s d s d d s d cd d

08 s d d s s d s s d s s d d d s d s s d d d

09 s d d d d d s d d d d s s d s s d s d d s

10 d d d d d d s d d d d d d s s s d s d d s

11 d d d d d d s d d d d d d d s s d d d d s

Key: The x and y axes are for Table 1 above. ± s and d indicate whether the informants thought that the words in question were the same (s) or different (d); bd or bs, used when informants were asked to identify three words as the `same' or `different' as in questions 1 and 7, indicates that only b was different or the same. ± ? signals either that the informant(s) were unable to produce an explicit answer, or that the investigator was unable to decide unambigously as to what answer the informant was opting for (e.g. (B p. 54 below ) and K: `Yeah both the same' L: `Yeah but one is more dogs and one is just one dog'). 48

3 What do Lay Speakers Say About Words?

Words and Linguistic Form The ®rst part of this chapter is an analysis of informants' perceptions of a word's identity in terms of `sameness' and `difference'. The second part is an investigation into the criteria used for identifying the word. This ordering is determined, in part, by the assumption that identity logically precedes identi®cation. The results of the empirical investigations seem to indicate that a word's identity is not determined by the invariance of phonetic or semantic features ± rather it depends on the point of view taken. But one needs to go further than the Saussurean axiom that it's the viewpoint adopted which creates the object1, and to reject the idea that there is one viewpoint from which to envisage a determinate identi®cation of linguistic items. Questions of identity are not related to the `same' words, i.e. words which have an orthographic constancy, i.e. the same spelling. Question No. 2 was an investigation into the criteria used in 1 `Other sciences are provided with objects of study given in advance, which are then examined from different points of view. Nothing like that is the case in linguistics. Suppose someone pronounces the French word nu (`naked'). At ®rst sight, one might think this would be an example of an independently given linguistic object. But more careful consideration reveals a series of three or four quite different things, depending on the viewpoint adopted. There is a sound, there is the expression of an idea, there is a derivative of Latin nudum, and so on. The object is not given in advance of the viewpoint: far from it. Rather, one might say that it is the viewpoint adopted which creates the object. Furthermore, there is nothing to tell us in advance whether one of these ways of looking at it is prior to or superior to any of the others' (Saussure 1922: 23)

49

What do Lay Speakers Say About Words?

establishing a word's identity. The results clearly demonstrate (Table (2) p. 48 above) that informants' responses were not solely based on orthography, dictionary entries, pronunciation, meaning or etymology. Instead they depended on individuals' interpretation of the question (which includes their interpretation of what they feel I wished to hear). One can even go further than this, showing how and what people say in their responses are often based on factors including attempts to impress, `pull rank' or question the questioner. I myself, for instance, found this latter point signi®cant when I was questioning my father, a tutor at my former university and my peer group during my graduate studies. In such cases my tutor appeared to be using as many technical words as possible, as was my father. My postgraduate peers appeared to be giving answers which ranged from a rejection of the question (if, for instance they thought linguistics was a nonsense study `this is where we get into nonsense') to showing off their newly-found knowledge. These factors are all integrated with such factors as educational and occupational background. As Table (2) shows, no one variable exclusively determines responses. Moreover, there was no conformity between individuals, and the majority of responses did not appear to follow any clear-cut procedures by which to determine word identity. There was thus very little consistency within the responses of individuals. Informant A was the only informant to volunteer the criteria which he consciously employed: A: I'm saying that they're the same because they're written the same and I'm saying that they're different because they're written differently but ± in the case of ± em ± er ± particularly pervert and pervert they're different pronunciation and they're still the same word but they mean different things as well Informant G also considered such words to be `the same', proffering reasons adduced by historical linguists to support his case: 50

What do Lay Speakers Say About Words?

G: I would say that the word bank is em a word which has a very wide semantic range I'm afraid I: Would you say it was the same word? G: Yes I would say it was the same word which has a very wide semantic range For the instances of pervert he said: G: . . . as a historical linguist I would say one is a noun and the next one is a verb em a but of course they're the same yes I: Why `of course' are they the same? G: What? I: Why are they the same ± you said `of course' they're the same' G: Em ± well historically they're just the same word that's all And for the words which were orthographically different he commented: G: I am a nurse I was a nurse yes in spite of the fact that they come from two different roots I still regard that as one word yeah Although a historical criterion is offered here, ascertaining the identity of diachronic units is even more problematic than ascertaining synchronic identities. But G here assumes that 1) diachronic identity is unproblematic, but 2) even so, he is not consistent, because he does not allow the different etymological histories of the various words bank to count. A number of informants introduced some discussion of meaning: a few thought that they were being asked whether the words were in some way synonymous: E: Do you mean do they mean the same? I: Are they the same word? Avoiding a direct answer here is a similar tactic to McGregor's: not telling the informants what criteria they were to use aimed 51

What do Lay Speakers Say About Words?

`not to prejudice the outcome of the research by providing information whose signi®cance one could not know in advance' (McGregor 1984: 74) (see also informant G p. 54 below). Basing identity on meaning was easier in the ®rst question, since explaining the difference between the sentences containing bank, is simpler than explaining those sentences containing the auxiliaries or the word nothing of question 2. N: I'd say they were the same M: No the ®rst one and the third one are places to keep things and the other one is a bit of ground N: That's true And K drew a distinction between the word and the meaning K: . . . the words are the same but the meanings aren't Although the doctoral candidates at Oxford University often backed up their arguments by appealing to meaning and word classes, they did not produce the same responses, or employ their criteria in the same way. For example, explaining why he thought the instances of nothing were the same in question 2, informant D said: D: I could think that you could exchange the two ± you could interchange the two ± and it would ± and it would mean the same thing Informant R however thought that these two units were different, whereas Q claimed that they were the same, the only difference being that of mention vs. use. Informants C, D, and Q said that pervert (n) and pervert (vb.) were different because one is a noun and the other a verb. Yet, despite the similarity of their educational background, their replies showed little conformity. Informant D, unlike informants C and Q, declared the auxiliaries in questions 6, 7 and 8 to be the same. And Q, referring to the words of different tenses declared them all to be the same, unlike the doctoral candidates C and D. Other comments elicited for these questions used appeals to (hypothetical) contexts: 52

What do Lay Speakers Say About Words?

I: N: I: M: N: I: M: N: M: N: M: N: I: M: N: I: N: M:

Are they the same words am and was? No they can't be Why not? Part of the same verb Same verb yeah that's true Are they the same word? Not in context no How d'you say they're the same word? Because ± They're different words They're in the same ± That would mean the same if it was now [pointing] Yeah But they're different words? Yeah Yeah But why aren't break and broke different words? Well they are but they're different contexts but it's the same thing They are but it's the same thing innit

Likewise E produced situational rather than abstract answers (see below). E: That's different innit I: Why? E: `Cause dogs is more than one . . . If you're a nurse you're a nurse ± if you leave the job you WAS a nurse I: So is it the same word or different word? E: Different According to Walter Ong, this mode of argumentation characterises an oral-based mode of thought and expression. `Oral cultures tend to use concepts in situational, operational frames of reference that are minimally abstract in the sense that remain close to the living human life world' (Ong 1982: 49). Often, my informants requested further information as to what was expected of them: 53

What do Lay Speakers Say About Words?

P: Is it different just `cause there's an s on the end, or is it the same [referring to question 5] I: The three words the same words or different words? [question 1] G: In what sense? I: Any sense ± what would you say? I: Are they the same words or different ± the two nothings? J: Do you mean do they mean the same? F: Hang on do you mean the meaning? I: Are the two banks the same words or different words? B: Well if you like they're the same words but they're being used in a totally different way I: So you think they're the same word? B: Well to look at them yes ± but it's a different descriptive thing so possibly it's two words ± I dunno ± what again what is a word? In two such cases, B above, referring to bank and L to dog vs. dogs (question 5), were unable to produce an explicit answer [signalled on Table (2) by a question mark]. The replies discussed above give grounds for questioning whether word-identity is as obvious and as clear-cut as linguists generally suppose (see the also the extracts from Sapir Chapter 1 above). Speakers of English have, on the whole, a fairly good intuition about what `a word' is, although there are a few marginal cases where intuitions are unclear . . . it could be pointed out that even preliterate children in our society have a reasonable notion where words begin and end . . . There is also a certain amount of evidence . . . that non-literate speakers of unwritten language know where words begin and end in their languages (Bauer 1988: 45). A similar problem occurs when theorists claim that went and has gone are `equivalent forms of the lexeme GO' (Bauer 1988: 9). 54

What do Lay Speakers Say About Words?

One is inclined to ask, as informant G above `equivalent in what sense'? Less than half of my informants declared questions 9, 10 and 11 to contain `equivalent' words ± and many used criteria other than those set out in grammar books. For question 9, nine informants declared the words to be the same, whereas for question 10 only six did, and only four questioned about number 11 thought that am and was were `the same'. Manifestly in Bauer's example there is no phonetic or orthographic equivalence; nor do the sentences have an equivalent meaning, if the criteria for that is `identity of descriptive meaning' two expressions have the same descriptive meaning (i.e. are descriptively synonymous) if, and only if, statements containing the one necessarily imply otherwise identical statements containing the other, and vice versa (Lyons 1981: 53). But `Jonathan has gone to Paris for four years' has not the same meaning as `Jonathan went to Paris for four years' : the one does not necessarily imply the other. Bauer's `equivalence' is not based on something we automatically, tacitly or subconsciously know, but on a pedagogic experience that we may or may not have had, or may have forgotten ± a point frequently acknowledged by my informants (see below). The Identi®cation of the word It was the purpose of Question No. 1 to ascertain how the `word' is identi®ed, both in speech and in writing (see Table (1) p. 47 above). The informants were ®rst asked to listen to the sentences and to state how many words they thought each sentence contained. When they had answered all the questions they were asked to read the sentences answering the same question. The questions were conducted in this way in the expectation that at least some of the informants might have forgotten their original answers. Answers elicited for other questions also proved fruitful for an analysis of word boundaries. 55

What do Lay Speakers Say About Words?

The Evidence Many linguists and psychologists apparently assume that `we all know what is meant by the word `word'' (Burgess 1975: 98).2 Accordingly, psycholinguistic examinations of word-recognition have focused on `lexical decision experiments' (Garnham 1985: 44) whereby subjects are asked to say, as quickly as they can, whether strings of letters such as ®lch, felch or welch, are words or not, or to pronounce strings of letters to ascertain whether there is a difference in the length of time they take between the reading of words and non-words. Few in this area of research have asked their informants to count words in a sentence, because it is assumed that all `normal' native speakers of English know, and agree upon, what a word is. But, as seen from Table (I) from my empirical work, only one question out of twenty-two, the spoken sentence 4 ± the Queen of England's palace is there ± produced a uniform response. The discrepancies cannot be explained in terms of high versus low levels of literacy alone, since there were marked differences even among the highly educated informants. David Barton's study of adults with a low level of literacy, although to be applauded for concentrating on the in¯uence of literacy on metalinguistic awareness, offers a far too simplistic explanation for the differences in counting words. Certain types of error turned up consistently: Typical errors were treating phrases with a unitary meaning, such as more or less, as one word. We refer to these errors as conventional errors. They are conventional in that an element of orthographic arbitrariness enters into the decision as to whether or not the form should be written as one word. There were very few examples of other errors of segmentation into words in our adult data (Barton 1985: 191±2). 2 `Indeed, of all the units of linguistic analysis, the word is the most familiar. In fact, its existence is taken for granted by most of us. We rarely have dif®culty picking out the words in a stream of speech sounds or deciding where to leave spaces when writing a sentence (O'Grady, Dobrovolsky and Aronoff 1989: 90).

56

What do Lay Speakers Say About Words?

Barton explains such `errors' by appealing to how linguists decide whether a sequence is a word. He claims the criteria they use involve `combining grammatical information with the fact of whether the form can occur freely in a sentence or whether it is always bound to another form' (ibid.: 192). The dif®culties that semi-literate adults face, he feels, are of the same kind as those faced by linguists in certain instances. The results, Barton concludes, `point to a qualitative difference between adults and children in the awareness of units' (ibid.: 192). He found that some errors made by children were not found in the adult data: These child errors include a refusal to accept noncontent words such as the as being real words, a speci®cation of a minimum number of letters in a word, and offering phrases instead of individual words as examples of words3 (ibid.: 193). In the present study, two adult informants committed so-called `child errors'. Informant O refused to accept anything smaller than two letters as a word: I: No I don't O: Two I: Where are the two words? ± what are the two words in no I don't? O: No and don't I: What about I? O: It's not a word is it, it's a letter I: O: I: O: I: O:

John's got a new car Four Where are the four words? John's got new car A's not a word? Well as far as I'm concerned it isn't.

3 My daughter Ella, aged 5.7 supports Barton's ®ndings above. She continues to insist that the, be and it are not words.

57

What do Lay Speakers Say About Words?

Informant E came to the same conclusion, but only when reading the sentences. On reading the sentence Have you got a colour TV? he deliberated as to whether a was a word: E: I'm just trying to be clever here and it don't work ± I wouldn't say a was a word I: You don't think a's a word E: No I: Why? E: No I don't think it is ± a's not a word is it? I: What is it then? E: It's a letter I: What about I then ± do ± E: I's not a word I: A letter? E: Yeah Both these informants insisted that the spoken sequence one, two, three, Go! contained only one word. But ®ve informants, when they had the written ®gures in front of them, denied that such ®gures were words: P: . . . 'cause they're not really words are they, they're just they're not actual words are they 1, 2, and 3 J, however, disagreed: J: Four words though you haven't got it written ± you've got 1, 2, and 3 in ®gures but one, two, three are actually words Informant M explicitly drew a distinction between the written and the spoken sentence: N: M: I: M: N:

Four em I don't know 1, 2, 3, WRITTEN written it's only one, spoken it's four Why is it only one written? Because those aren't words, those are numbers Yeah but if they were written as words it would be four wouldn't it. It depends how you look at it 58

What do Lay Speakers Say About Words?

M: You call them letters and ®gures N: But because they've got commas after them they're each a separate thing Informants E, O, P and T said that ABC was not a word and therefore the utterance Do you know your ABC? contained only four words: P: ABC aren't words they're letters Informant J, although also denying that the sequence ABC was a word, used a different method of counting: I: J: I: J:

What is it? Well it's three capital letters So how many words would you say were in that sentence? Do you know your ABC? Do you know your ± I should imagine seven I: So the capital letters are individually one word you're saying? J: Yes When this informant commented on the same written sentence, he said that `ABC was an acronym for the alphabet', and the sentence contained four words since in itself ABC was not a word. This analysis followed his discussion of the sequence TV, which he denied to be word. I: What is it? J: It's a ± what do they call it ± a a metaphor ± it's not a metaphor ± I: But it's not a word? J: It's not a word ± it's it's a it's em er letters that describe words like radar or that describes words I: So you would say radar wasn't a word? J: It's not a word ± it's it's a it's em what do they call it now? ± though it's used it's not actually a word ± it mean em ± some some it it's each each ± the ®rst letter is ± capital letters mean `radio' and something 59

What do Lay Speakers Say About Words?

I: J: I: J:

Okay What is the word? Acronym? Acronym

Although informants N and M agreed with J's analysis of ABC `ABC's only a way of saying alphabet' they disagreed with him and among themselves as to how many words the sentence contained: M: I'd put seven N: I'd say ®ve Yet this pair, after some deliberation, agreed with each other, and with their original count, on the sentence containing TV: M: TV bugs me N: Yeah it bugs me, I'd say six `cause TV's only an abbreviation for television M: Television's one word yeah Informant E, adhering to his exclusion of a as a word declared that this written sentence contained four words: I: E: I: E:

Why four? Well TV aint a word is it? Why? Abbreviation for a word

Informant O claimed that the spoken utterance contained ®ve words because a was not a word, but the written form contained four. Informants K and L thought that TV was two words L: K: I: L: I:

That could be six or seven Seven What do you say? Six And you say seven? 60

What do Lay Speakers Say About Words?

L: K: L: K:

Well TV is sort of one word but it's two letters Yeah But that counts as one word ± it's still the same as television Yeah seven

Informant O, just as he was excluding single letters from the status of a word, also denied I've to be a word in the written sequence in question 3: O: I: O: I: O:

Well you can't count that `i-v-e' You can't? No I didn't I'm sure I didn't Alright so it's just got to go to London ± you think? Yeah, and that totally alters the sentence don't it

B was the only one who considered the three units don't, I've and John's to comprise two words in the spoken as well as the written sentences: B: You trying to count an apostrophe as one word or do you want it counted as two? I: What do you think? B: Well as far as I'm concerned it should be John has got I: So if you see John's or say John's you'd ± you'd say it was two words? B: Yeah ± I'd think of it as being two words ± well that's what I was taught anyway Informant M was the only other subject, despite objections from N, who concurred with B in claiming that the orthographic don't was two words: M: I'd still call that four N: That is it it's hard for that because of the apostrophe, it makes it one word that's the whole reason you put the apostrophe in to stop ± so you don't have to write do not which is two words 61

What do Lay Speakers Say About Words?

The general linguist Q was the only one who considered the `s in the written sentence of question 4 ± the Queen of England's palace is there ± to be a word. Attitudes to the representation of hyphenated, solid or open compounds varied, as is seen by the responses to questions 1, 6 and 8. D, E, W and M thought that windsur®ng, when spoken, consisted of two words. Yet on reading this sentence, there was total consensus that it was only one word. Washing-machine however, caused a disagreement: N: I'd say six `cause they're separate, they can be separate words M: But then again they're hyphenated N: They're hyphenated but you could use washing on its own or machine on its own so I'd say it was two words M: I disagree G was the only informant to declare hide and seek to be one word: G: Let's play hide and seek em that as far as I'm concerned is three I: Why? G: Because hide and seek is one word I: What's your de®nition of a word there? G: Well hide and seek should be hyphenated er because it's it's it's em it's a word which is altogether thing you know what you know like cara- no not caravan em I can't think of anything em but hide and seek can't be three words surely can't can it, hide and seek got to be one word it's a it's a one thing er that's my feeling you know er you don't need to agree but er as far as I'm concerned it's three words hide and seek is a word At this point it is illuminating to recall Barton's discussion (p. 56 above) of `typical errors' in particular `treating phrases with a unitary meaning, such as more or less as one word' (Barton 1985: 191). These errors were attributed to adults of a low level of 62

What do Lay Speakers Say About Words?

literacy. My informant here was a historical linguist and a Reader in Middle English at London University. From the answers elicited from the other tests, it was clear that word boundaries were not only drawn on the basis of phonetic, phonological, semantic, orthographic criteria or lexicographical practice. G, for example, claimed that imbloodypossible was `one of those strange words made up by somebody and therefore one word which can't be analysed in anything else'. T thought that when spoken the sentence contained either three or ®ve words, but when written three; whereas W and H both claimed that the sentence contained six words. Both O and J declared that the written sentence contained only three words because imbloodypossible was not a word: O: that aint never a word that imbloodypossible I: It's not a word? O: No impossible's a word without the bloody in it J: I don't think there's such a word as imbloodypossible as one word [sic!] However, when J heard this sentence he provided a different analysis claiming that there were ®ve words ± no it is bloody and impossible. This contrasted with O who said that bloody was not a word: O: it's not in the dictionary is it? ± it's a swearword innit as far as I'm concerned so it would be im ± no it's bloody I wouldn't count bloody as a word it's just a slang ± it's not a word This rejection of bloody as a word coincided with O's view of swearwords in general I: O: I: O: I:

You said earlier that em bloody wasn't a word No Do you think blimey and shit are words? No not really they're just slang words as far as I'm concerned They are words or they're not words? 63

What do Lay Speakers Say About Words?

O: Oh they are words but they're not words that you get taught in school if you know what I mean I wouldn't er if a kid of mine said shit or bloody I think'd have a go at it, blimey not so . . . there are words and there are words, and some words you can use in ladies' company and some words you can't, and the words you can't use in ladies' company are obviously to my mind not words For this informant, as for many of the others, perceived lexicographical practice was a determining factor in identifying a word. Informant V said that ouch and hey were not `proper words' (Question No. 3): they were exclamations that you would say `if someone hurt you', and `you wouldn't ®nd them in a dictionary'. A said: A: I would term it not to be a word or think it not a word if it wasn't in the dictionary Many of my informants denied that em, Question No. 4, was a word: P: I wouldn't say em's a real word it's just sort of like something you do, a noise it's more like a noise than a word O: . . . as you say isn't a word it's just a grunt if you like, I wouldn't have thought it's a word Q: Nothing, not a word S: No such word H, in claiming that it was not a word, was either inconsistent in his explanation or was using two different senses of word (as was J p. 63 above): H: Well not it's for formal writing but it is I mean people talk like that I: Would it be a word in speech? H: I don't think so it's just the way words [sic] come out sometimes, we don't know how to say it so we ®ll in the space with an em 64

What do Lay Speakers Say About Words?

And G, even though he thought that there was a technical term for such a linguistic phenomenon, still categorically denied it to be a word: G: . . . there is a special phrase for that which I've forgotten you'll have to supply that . . . no it isn't er it isn't a word of course not no no This contrasted with E who, after some deliberation, seemed to imply that in certain situations, em could be considered a word: E: I: E: I: E:

It's not even a word is it? Why? `It's just a sound Why isn't it a word? You don't look at a book and start reading em em do you ± unless you're listening to a Matteson's Advert4

E's comment contrasted with V's who said: `it's incorrect . . . you say it when you're not thinking . . . you would say it but never write it' and with T's who said that it was . . . : `for how you want someone to say it, written in a script'. For the highly educated informant D, it was clear from his de®nition that words were conceived as primarily oral: D: A word is any one of a number of different units used orally by human beings to convey mental ideas to other human beings This view of the word led him to deny em the status of a word: D: Well it's not really a word at all I: Why? D: Because it represents a sound ± er ± yeah it represents a sound and not an object or an abstract idea

4 This was a popular television commercial which used the phrase Mmm Mattesons and the phrase was subsequently printed on the products

65

What do Lay Speakers Say About Words?

The `Morpheme' Many of the problems that have confronted theorists have arisen because the criteria they have used for de®ning the word ± semanticity, determinacy of form, etc. ± also de®ne elements smaller and larger than the word. In the course of this present research, many informants proved capable of analysing words into smaller elements and assigning some sort of interpretation to such elements. For G -eth in witnesseth (Question No. 3) `means sixteenth century language': G: witnesseth has got to be a biblical term because it's got the `e-t-h' on it whereas for A this particular word is biblical because A: the -seth on the end of witnesseth makes it more into that N notes that ontogenetically is scienti®c because `anything that ends in -ically usually is a science': whereas A and F pick out genetically and B -genetic. In epistemological A comments on the -logical part, F on the -gical and J on the -ological. When asked to comment on hereinafter D comments D: the running together of three words ± em ± in a kind of formula that's very common in legal documents Even when talking about em two informants referred to the spelling: C: Well you'd spell that `u-m' ± you always spell that `u-m' M: I: M: N:

Oh that's a sort of little thought-thing Is it a word? No that's what you'd see spelt in so many different ways No it's a noise it's like um yeah `cause you could put `u-m' there couldn't you instead of em

66

What do Lay Speakers Say About Words?

Thus the criteria employed for the identi®cation and identity of the word have been shown to differ markedly and, in so doing, have called into question other linguistic units and categories as identi®ed by orthodox linguistics. For example, for informant A (p. 50 above), it appears that the orthography of a linguistic unit determines sameness and therefore questions of homonymy and polysemy5 are irrelevant. Although informant G could also be said not to recognise such considerations by declaring all the words within each question to be the `same' (pp. 50±51 above), he also seems to ignore the difference between what is known as the type-token relation6 (Hutton 1990). Q, the linguist, unlike C and D, seemed to be appealing to the linguistic concept `lexeme'7 when he declared the words of different tenses in questions 9, 10 and 11 to be the same. The responses to such questions also highlight the difference between informants' perceptions of the written and spoken word. Although both E and O denied that the sequence one, two, three, written in ®gures or heard, were words, others thought them not to be words only when written in ®gures (p. 58 above). Throughout the interview, E continually emphasised books and reading when talking about the `word'. Thus I would suggest that, for him, words were primarily written units of little concern to him (see his comments on em p. 65 above). The same applies to O's denial that I've was a word (p. 61 above) since he also claims that the omission of this unit in the sentence I've got to go to London `totally alters the sentence'; the concept `word' is apparently not necessary for communication and therefore of little import for this informant. There seems to be a gap between O's perception of what is necessary for communication and his de®nition of `word'. 5 Polysemy is usually taken to refer to one word which has different meanings (such as pupil ± student and part of the eye) whereas homonymy usually refers to two different words which have the same linguistic form (bank ± of river and ®nancial institution). The ®rst example is said to be polysemic because both meanings of pupil have a historical relationship. 6 This relation is said to be the relation between an abstraction of some kind ± a type ± and the realization of that abstraction ± the token such as between a `word' and its instances. 7 Lexemes are `the expressions that one would expect to ®nd listed in a dictionary' (Lyons 1981).

67

What do Lay Speakers Say About Words?

Even the diacritics used in writing do not seem to provide for an unambiguous identi®cation of linguistic units. Although hide and seek, question 8 of Question Number One, was written neither as a solid nor hyphenated compound, the spoken phrase caused the informants to label it as one word, three words or one or three words (T). G was the only one who said that it was one written word, thus appearing to base his judgement here on semantic criteria. His use of semantic criteria was borne out in his opinion that washing-machine and TV were one word. Some informants appeared not to draw a distinction between the morpheme and word. For example in the claim that imbloodypossible contained three words (O and J p. 63 above). Others (p. 64 above) generally denied as a word the phenomenon identi®ed by Howard: `The most frequent words will be the two most popular words in the English language, the ones that never get recorded in the dictionaries er and um' (Howard 1984: 193). These results call into question statements claiming that young children's understanding of the concept word differs signi®cantly from adults' and that, again, adults have a more complete conception of word as a unit' (Barton 1985: 193). Moreover, such investigations call into question the telementational model of communication (p. 5 above) which postulates a match between form and meaning, and moreover requires the match to be exactly the same for all members of a given speechcommunity. According to this model morphemic boundaries have to be drawn by all speakers in the same places; no phoneme can be unaccounted for, every morpheme has to be assigned its characteristic form or forms, and each such form has to be assigned a consistent meaning. From the above discussions one can see that this model is doubly defective in that there is little, if any, consensus on the demarcation and interpretation of the `word' and yet, despite and because of the lack of explicit instruction as to what a word is, all the above informants had strong opinions on the topic. One could conclude that such a concept is `psychologically real' in the 68

What do Lay Speakers Say About Words?

Saussurean sense8, without Saussure's stipulation that the `reality' is the same for all members of the same language community. In other words, although all informants are familiar with the term word and accept it a part of their language; they all understand it differently, some even (p. 64 above) using it in at least two different senses. The Spelling and Pronunciation of Words Part of the lay concept of words is that they can be spelt and, if spelt correctly, can be read and pronounced. The relationship between spelling, reading and pronouncing is intricate and highly sophisticated. In present-day British culture, one usually talks of spelling words, not phrases or morphemes, and the `tool' of spelling is the alphabet which is considered to be a representation of sounds. In OED2 for example, `spell' is de®ned as `To form words by means of letters; to repeat or set down the letters of words; to read off the separate letters forming a word or words' [my emphasis]. Accordingly, the `ideal' alphabet will be one in which each letter represents one and only one sound. In fact, Sapir seemed to think that there was, in practice, such a correspondence: each element (letter or written word) in the system corresponds to a speci®c element (sound or sound-group or spoken word) in the primary system. Written language is thus a point-to-point equivalence, to borrow a mathematical phrase, to its spoken counterpart (Sapir 1921: 19). Yet, as Saussure pointed out, there was a disparity between a word's spelling and pronunciation. Why is it that in French we write mais and fais, but pronounce these words me and feÂ? Why does the letter c in French often 8 `In order to determine to what extent something is a reality, it is necessary and also suf®cient to ®nd out to what extent it exists as far as the language users are concerned' (Saussure 1922: 128).

69

What do Lay Speakers Say About Words?

have the value of s? In both cases, French has kept spellings which no longer have any rationale (Saussure 1922: 49). Saussure's conclusion was that `writing obscures our view of the language' (ibid.: 51) and, because of this `tyranny of the written form' (ibid.: 53), `the more inadequately writing represents what it ought to represent, the stronger is the tendency to give it priority over the spoken language' (ibid.: 52). For Saussure, written signs were signs in their own right; the purpose of writing is to represent speech. Although Saussure correctly identi®es the mistaken assumption about the relationship between letters and sounds, he too was guilty of a related confusion. As Roy Harris comments, Saussure's assumption that speech comprises a linear sequence of discrete sounds was itself an extrapolation from the familiar structure of the written word. The notion that in speaking we select the individual consonants and vowels which somehow emerge from our mouths threaded in the right order like beads on a string is simply the image of alphabetic orthography projected back on to speech production (Harris 1986: 41). Questions Nos 8 and 9, an investigation into the spelling of certain words, addressed the relationship between letters and sounds. A number of opinions on questions 2, 3 and 4 (Question No. 8), were that the spelling was wrong because the word was spelt how it sounds. P: color is spelt the way it sounds . . . realise, the second one with a `z' that's more likely to be the way of spelling it, that's how it sounds it sounds as if it's got a `z' in it I: So you think that probably the b) realize one is more correct than the other P: No the one with the `s' is probably more correct, the one with the `z' is how it sounds I: So a correct spelling is not how it sounds? P: No de®nitely not I: He's a shef at the Ritz P: That's supposed to be a `c' instead of an `s', that's how it sounds 70

What do Lay Speakers Say About Words?

Informant E shared this view, although he spoke of the reading rather than the spelling of words: E: Well it reads colour dunnit ± when you look at it but you know it's wrong I: Okay E: Yeah I: Why is it wrong? E: Well it's got a letter missin' innit ± they look like they're spelt the way you say `em . . . Now he's a shef at the Ritz ± it's funny ± that's the way I'd spell chef ± yeah I: That's the way you spell it? E: Yeah ± that's the way I would spell it I: How do you know it's not right ± or ± E: Well I don't ± it's the way it sounds I'd spell it like . . . Well maybe if I was looking at it ± if I was reading it I'd just read it ± maybe more out of memory than looking at it properly I: So you think it looks okay? E: Yeah ± it looks Alright to me ± probably aint ± elizabeth don't look too right but ± He also thought that right in question 5 was spelled correctly: E: That's Alright ± there's nothing wrong in that ± I can't see no spellin' mistakes in there Hayley Another informant invoked phonetic considerations, claiming that the American way of spelling certain words was phonetic, in contrast with the British convention. J: What color is that ± well color is spelt without the `u' ± and there's a `u' in `colour' as far as we're concerned ± of course the Americans don't spell it that way ± phonetically I: So we don't spell it phonetically? J: We don't spell it ± I: But that is a phonetic spelling would you say? J: Yes 71

What do Lay Speakers Say About Words?

I: Is it incorrect to spell it like that if you're English? J: Yes According to D, a word spelt phonetically does not provide for ease of reading, as seen in his comments on luv and color: I: Are they both okay to spell like that? D: I would have thought so that in the ®rst case no and in the second case yes I: Why? D: Em ± perhaps because ± perhaps because the the er ± transformation of the word in the ®rst changes the original word so much whereas in the second the omission of the the letter doesn't change it ± em ± in the sense it still looks like it ± it's still identi®able whereas in the ®rst it's not ± it's almost like ± it's almost like em ± attempting to read a sentence which is made up of words spelt in er ± phonetic symbols em ± I couldn't accept that either ± I'd ®nd it dif®cult to read A number of informants said that the spelling of questions 1 and 6 indicated how the words were to be said: A: That's how you would write it in em ± say a play to describe how you would want the word said I: So how ± A: Luv ± you ± you would want it spoken as a Northerner or as a Cockney ± so it's describing how you would want it spoken . . . it's the way you would write it in a song or ± or er ± a play saying how you wanted it to be spelt ± in other words you're dropping the `g' M: . . . you wouldn't want somebody who's actually singing to say `I'm sing ± ing ± [g] in the rain', it would come out I'm singin' in the rain J: I: J: I: J:

Love is not spelled that way except in slang terms So that's a slang spelling? Yeah or pronunciation the slang How would you pronounce it differently from reading ± [luv] 72

What do Lay Speakers Say About Words?

I: As opposed to? J: [l6v] . . . . . . Well it's it it's it could be somebody actually actually ACTUALLY singing and not mis- not pronouncing letters [singin'] H: singin' should have a `g' on it but since we have our little apostrophe I guess for writing so we will read it how it comes out of our mouths and we'll say [siÎin] instead of [siÎgiÎ] When informants were asked to state the justi®cation for `correct' spellings, opinions diverged widely. Informants not only appealed to the `representational' character of letters but also to pedagogic instruction, language-speci®c practices and certain culture-speci®c artefacts such as a spelling check on computers and conventions of certain printing houses. M: Um realize I use `s' N: I use the ®rst one yeah I use `s' M: Again that's how I was taught N: They both make sense I wouldn't sort of not know what one of them meant M: Then again I ± I: Do you think they're both correct or ± M: No the `z'strikes me as wrong ± the whole purpose of having the `s' on the end with the `e' on the end is to make the `s' sound like it does N: You make the `s' sound like a `z' if they put the `z±e' J: I: J: I: J:

. . . well we spell it with an `s' So the `z' is wrong? As far as the English is concerned Well who spells it with a `z'? Again maybe Americans

A agreed with J's analysis both of color (p. 71 above) and of realize: A: color is spelt wrongly . . . realize should be spelt with an `s' and not a `z' 73

What do Lay Speakers Say About Words?

I: So the `z' is wrong? A: Yeah ± yeah ± I believe the Americans spell it that way don't they I: But it's wrong? A: It's wrong N, M and O claimed that the American way of spelling certain words had in®ltrated into England as evidenced in books and on the television O: What color is that? well that's the way they spell colour nowadays you only gotta look in the TV that's how they spell it without the `u' so I presumably one is the English way and one is the American way M: They can't seem to make up their minds which way round it's spelt, sometimes it's `o-u-r' sometimes it's `o-r' N: It started off as `o-u-r' and now we're getting more Americanised M: Well I stick to the `o-u-r' `cause that's how I was taught N: Yeah I do ± ± ± but you get an awful lot of books now I mean not necessarily from American authors that are spelt like that F, however, was more liberal in her judgements: I: Is it wrong to spell it that way? [color] F: Well I suppose it depends on the way where you are and the way you were taught ± it's alien to me but I can understand it when I see it written . . . They're both two right [realize] I: They're both F: They're both quite good words realize G claimed that both instances of realize were acceptable, the only difference being that of printing house conventions: G: We're talking about the `s' and the `z' aren't we em I don't think I feel very strongly about those at all and I can never remember which to do but certain printing houses have certain practices 74

What do Lay Speakers Say About Words?

H, the American, had a rather complex way of deciding what is correct and incorrect spelling, relating such distinctions to grammaticality. He was not in agreement with the other informants on their distinctions between the American and British way of spelling: H: Well luv is spelled `l-o-v-e' I: Is that wrong? H: Well it's I don't think it's proper grammar but it's you still get the meaning across ± it's easy to write it that way I guess less letters to make your hand write I: Would you say it was wrong? H: I would not spell it like that on a paper I had to turn in because the teacher would tell me it's wrong and take points off so ± I: Okay H: Other than that I don't see anything wrong with it ± em what color is that? That's how I always spell color but it's also spelt `c-o-l-o-u-r' I: Which one is right? H: I think they're both right em I think both are accepted as proper spelling of colour . . . . . . realize I would say the `z' is correct but I don't know, if I didn't have a word-check on my computer I'd be really hurtin' so I would say the `z' is correct if I had to guess I: And why did you not want to say whether color was correct or not, how can you say that realize has got a correct spelling but color hasn't? H: Well I've seen color spelt both ways . . . I don't think realize is spelt both with an `s' and a `z' and accepted both ways . . . . . . I think one of the two is correct and the other one is wrong but and I I I'll say the `z' is correct I: Do you always spell it with a `z'? H: I think so see I write when I write my papers on the computer then it tells me which words I spelt wrong . . . . . . I would say shef should be `c-h-e-f' I: So is that wrong? 75

What do Lay Speakers Say About Words?

H: Yes it's spelt incorrectly yes that's wrong . . . I: Well you didn't say that about the ®rst one about luv H: Well I have seen it spelt like that but I don't I don't I still don't think luv is correct ± grammatically correct I: Is it wrong? . . . H: I don't think it's wrong but I'd I wouldn't do it like that because other people do I: But number four you think is wrong? H: Yes I: So your criterion for something being wrong or right is whether you've seen it before so if you saw chef spelt with an `s' you would say you would no longer say that it is wrong because you've se- em seen em seen luv spelt with a `u'? H: Well that's probably a bad way to decide what's right and wrong but that's pretty much what I'm doing yes you're right When asked to respond to the questions on Question No. 9, a number of informants declared letters to have more than one sound corresponding to them. A claimed that one is taught, when learning to read, that some letters have two sounds: I: Does mat and cat ± the ®rst word the ®rst letter in each one differ in any way from say the word at or art ± the ®rst letters? A: Yeah yeah ± em cat going back to early schooldays you would think that that began with a `k' ± for that's the sound that it is ± whereas at is de®nitely an `a' sound ± like you know ± once again going back to early schooldays when you're in the infant school that's the way that you learn to read and write is that you ®rst of all you learn the sounds of certain letters and ®nd that some have got two sounds and ± I: Does ± does do sounds say ± the `k' sound in cat as you said or the `m' in mat differ from the a at the beginning of a word or an e at the beginning of a word? A: Well well that's what I have just said ± but it ± it a ± to say `a' at the beginning of a word ± you couldn't have two sounds ± or I suppose `c' can have two sounds couldn't it? sound like an `s' or say like a `k' ± `m' and `c' are consonants and `a' and `e' are vowels 76

What do Lay Speakers Say About Words?

Other informants claimed that, instead of a letter having more than one sound, the presence of a letter can modify the sound of another: I: What's the difference between `c' `m' `a' and `e'? N: Well `a' and `e' are ± M: They sound different on their own . . . . . . The letters on their own sound different from each other, it's when you use them with another letter you get a different sound, you've got [ñ] and you've got [k] [in cat] but if you had it with an `e' it would be pronounced completely differently I: Why do they all sound the same [the initial sounds in psychology cycle and sight] B: I'll never be able to answer the one that begins with a `p' because I've never been able to understand that myself em ± `c-y' begins er ± the `y' makes the `c' soft and `s' is pronounced soft all the time anyway G was the only informant to distinguish between a written and a spoken consonant: he declared the `same' consonant can yet display phonetic difference: I: Does the word cat begin with the same consonant as the word kill? G: Em yes it does yes they are both consonants I: The same consonants cat and kill? G: Yes they are yes yes they're the same consonant though they have an em different em phonetic phonetically they're different phonetically I: Phonetically? G: Yes cat and kill both begin with the same consonant but em phonetically they're slightly different I: If you read cat the word cat and kill would you still say they were then the same consonant? G: Yes er em do you mean speaking or writing? I: Writing G: Oh writing no of course they're not no ± speaking they are I: Speaking they're the same consonant writing they're not? G: That's right yep yep 77

What do Lay Speakers Say About Words?

Although for G word-identity was unchangeable irrespective of the medium in which the word was realised (see Table (I) p. 47 above), this was obviously not the case with the identity of consonants. This contrasted with A who agreed with the proposition that he was using the term consonant to apply to a written unit: I: Right ± does the word cat begin with the same consonant as the word kill? A: No I: Why? A: Kill is a `k' ± and cat is a `c' I: So a consonant is how you write it? A: Yes Unlike the other informants, G claimed that the words sheep and dumb both contained two consonants and one vowel; my other subjects tended to say that these words contained three consonants and two vowels and three consonants and one vowel respectively. P volunteered the information that sheep contained two vowels ± `well there's one vowel but repeated'. The fact that units below the level of the word, in this case letters, can be quanti®ed, plays a key role in the lay person's concept of linguistic form as seen in the assignment of terms applied to antidsestablishmentarianism and it in question 1 of Question 3. Although fourteen informants declared that the ®rst word was `long', other terms used and the comments on such categorisations were extremely diverse. Informants K and E did not use the term `long', instead they appealed to the number of letters in this particular word: K: Well there's more words letters in antidisestablishmentarianism than there is in it E used the term `big' and not `long': E: I know how to say it and I don't know how to say the other one ± I wouldn't bother with the other one I: Anything odd about the other one? E: What the big word? 78

What do Lay Speakers Say About Words?

I: E: I: E:

The big word No Why is it a big word? `Cause it is ± how many letters has it got? [rhetorical question]

Informant F used the terms `big' and `long' relating them to the word's meaning: F: Em the ®rst one's sort of bringing more meaning to something ± and it refers to nothing really I: Anything else? ± about the ®rst one F: It's a big word I: What d'you mean by a big word? ± big in what way? F: It's long and it can mean it's got a lot more meaning to it and you'd have to dissect it to ®nd out N and M suggested a reason why such a word was `complicated': N: Yeah well it would confuse a lot of people wouldn't it I mean look at the size of it ± it is THE longest word isn't it it must be something like that M: It's a complicated word I: What do you mean by a complicated word? M: Well if you said to somebody `spell that' nobody would be able to most people wouldn't be able to N: I mean I recognise that therefore I know how to say it N: Yeah I do M: But somebody looking at that for the ®rst time would have to read through it a few times to work out how it's pronounced N: You'd have to go through it a few times saying it in different sections to get it right wouldn't you Question 11 of Question No. 3, tsetse and phthisis, according to some of the informants, also created pronunciation dif®culties; this time the problem was not related to the length of the words: 79

What do Lay Speakers Say About Words?

P: It doesn't look like a real word I: Why? P: Can't think of how to say it C: I have a slight lisp so I hate seeing words like the second one I don't know if you count the `t±s' as a diphthong but obviously it looks like ± the fact that you've got two consonants together or two double sets of consonants in the second one ± em ± well you don't normally see them in words makes it dif®cult to say so you just look at both of them and think fuck how do I pronounce this? M: That's a weird word I: What's weird about it? M: It's an unusual combination of letters N: It's got ± yeah you normally only get one `p-h' or `t-h' in a word not both M: Then a really simple ending bit on it D: Well there's nothing necessarily odd about them but they em ± they just create a kind of ± em create a kind of mental idea that they are dif®cult to pronounce ± especially the `t' and the `s' ± the recurring `t' and `s' in the ®rst word ± em to ®nd it in the middle of the word and not at the end is probably a little confusing ± em ± the other the other word contains the problem for me of having the `t' and `h' placed ± er ± straight together Some were sceptical about whether phthisis was in fact a word: J: Well one I know is a ¯y ± well the second word I've never seen a word start with four consonants I: Okay J: Is it a word? W: you can't HAVE the second word Informant G produced arguments for both being `possible' words: I: How do you know the second one is a word? 80

What do Lay Speakers Say About Words?

G: How do I know it's a word? I: Mm G: Em well of course I don't in a sense you could have made it up um `p-h-t-h-i-s-i-s' well `p-h-t-h' em is a possible combination er unusual in English but possible if it's a loan word The logical extension of this argument is obvious; if a combination of consonants is `possible' because the word in question is `a loan word' then, because there might be no limit to the combination of consonant sequences in foreign words, any series of consonants is a `possible' word. Informants A and O referred to the fact that these words contained `silent' letters, but the letters they identi®ed as `silent' were different in the same words A: It's got a silent `s' and [®…is] got a silent `p' sound as if an `f' ± they've both got silent letters O: I presume the `t' is silent so its tsetse and phthisis same thing Barton comments that there is in fact `no reason to think of words as having a single representation' (Barton 1985: 203). He conducted an experiment asking adults of various literacy levels to say the ®rst sounds in trouble and other words beginning with tr-. He found that four different responses were given: t, tr, t , and tSr: `Adults who could spell at least the beginning parts of these words would give any of these four sounds as their response, but the few who could not spell the beginnings would never give t as a response' (ibid.: 196). Incorporating the results of Read's investigations (1975) that tr- was frequently spelled ch(r)- by children who taught themselves to write, Barton concludes: `If we interpret all these results together, then it seems plausible that it is only as a result of seeing tr- words written down that people analyse them as being composed of t plus r' (ibid.: 197). A nonliterate `does not have a representation of train where the ®rst sound is t'; that, he says, `comes from experience with orthography' (ibid.). 81

What do Lay Speakers Say About Words?

Certain phonetically identical words, as Sapir noted, are perceived by non-linguists to be different, as in soared and sawed. Among educated but linguistically untrained people who discuss such matters differences of orthography are always held responsible for these differences of feeling. This is undoubtedly a fallacy, at least for the great mass of people, and puts the cart before the horse. Were English not a written language, the con®guratively determined phonologic difference between such doublets as sawed and soared would still be `heard' as a collective illusion, as a true phonetic difference (Sapir 1949b: 54). Bolinger expanded upon this discussion in his attempt to show that reading necessitates an integration of eye-movements which plays a non-trivial part in a word's signi®cance. If one believes words only induce `laryngeal' movements, then the fact that certain homonyms are more dif®cult to elicit than others remains inexplicable. Bolinger required thirty-four students to produce a homonym for each of the words in the phrase I bear no sword here for thee O great one As predicted, the signi®cant ®gure is for soared. These tests prove that, other things being equal, the more unlike the spellings are, the less likely it is that the identi®cation will be made. For all the fallibility of self-observation, it is well to remember that one can FEEL a difference of eye movements in passing from beaut to butte, from key to quay, and from rapt to wrapped (Bolinger 1946: 336).9 Moreover, differences in the spelling of the `same' word often cause a different attitude towards the words concerned. Bolinger 9 Although in Western literate society, soared may well be signi®cant, it is only signi®cant in a statistical sense. For example, the ®rst person I gave Bolinger's test to, was an English Professor at London University. This Professor was unable to come up with a homonym for thee within an allocated time scale. When he asked for `the answer' I spelt out the word `t-h-e'. His response was `oh that's […@], I always say […@]

82

What do Lay Speakers Say About Words?

found that of the pair gray and grey, grey is `preferred for the good, gray for the bad' (ibid.) and my informants N and M commented on the variant spellings of grey and hello in the context of their discussion of the in¯uence of Americanised spellings (see p. 74 above) M: Then again you get the same with grey with `g-r-e-y' or `g-r-a-y' N: Yeah I mean look at how many ways you can spell hello I mean that's another one innit Bolinger also cites instances of dialectal spellings, or `eye dialect', which not only show a mispronunciation but also `burlesque the words or their speaker' (ibid.: 337). Since /'wõmõn/ is a standard pronunciation of women, the corresponding spelling wimmin is eye dialect. The spelling here is a visual morpheme which implies, `The person quoted is one who would use a vulgar pronunciation if there were one' . . . Misspelling is made to substitute for mispronunciation: i.e. it is suggested that the speaker is at the level of ignorance where one misspells in this fashion, hence mispronounces as well (ibid.: 337). Informants' attitudes to question 1 (Question No. 8) supported Bolinger's view that visual morphemes `exist at their own level, independently of vocal-auditory morphemes' (ibid.: 340). However, contrary to the implications behind Bolinger's analysis, my informants' reactions were not uniform. D, E and M thought that luv was a `slang spelling' but all gave different justi®cations: M: It's a slang spelling `cause it's different D: Luv is er ± a kind of slang abbreviation arising from the way that the word luv is pronounced E: I dunno how this way of spelling love came about in the ®rst place ± is it just a slang word ± a slang spelling? Others evaluated the sound and the `register': 83

What do Lay Speakers Say About Words?

N: It's one of those things that makes them sound cute F: It's like a comedy sort of `love' ± the word luv is a term of endearment A similar opinion on this phenomenon is from an Oxford tutorial fellow (p.c.): `I sometimes get notes from students concluding luv etc. I would be disturbed if they wrote love, but take luv as a not too serious gesture of friendly informality. A rather vulgar ªmateynessº'. B, however, commented factually on the `correct' spelling: B: luv should be spelt `l-o-v-e' not `l-u-v' H referred to the `grammar' of the word luv (see p. 76 above) and G concurred with Bolinger that such a spelling indicated the low standard of the speaker's education. Unlike F, N and the Oxford tutorial fellow, he could see no difference in terms of the meaning: G: . . . these are what I called in the recent book I've written empty spellings em they're spelling which are er supposed to indicate that the speaker is, so to speak, subliterate or something like that er there is no point in it at all, absolutely useless er because `l-u-v' means no different from `l-o-v-e' I: Do you approve of these spellings? G: Well I can't ± they're called eye ± they're called eye spellings I think by the Americans or eye what eye rhymes no no eye spellings I think er in dialect writing I I call them empty spellings er there's no point of er er because `l-u-v' means exactly the same as `l-o-v-e' The majority claimed that shef and right (Questions 4 and 5) were `wrong' but what was identi®ed as `incorrect'differed in each case: J: Well that's completely wrong I: Is it worse than chef spelt `s-h-e-f'? J: Yeah `cause it's just a letter in shef it's just one letter that's wrong but here the whole word is wrong 84

What do Lay Speakers Say About Words?

D thought that shef and luv were unacceptable: D: the `s' beginning the word shef is unacceptable because it doesn't em ± it doesn't em ± it doesn't ± er ± acknowledge the French origin of the word the way that `c' does I: Why isn't color unacceptable or luv unacceptable? D: Yeah ± well luv is unacceptable ± color because the omission of the single letter in color doesn't ± em ± you know aesthetically change the word as the case of `s' in shef But for Question 5, he did not refer to a spelling mistake but rather to an `incorrect' word: D: . . . the incorrect verb there ± em ± confusion perhaps arises from the fact that right as used here em ±is pronounced in exactly the same way as the different verb `write' is Words written in lower case letters (Question 8) also caused some informants to react to the `incorrectness': G: . . . that's without the capitals is it? Yeah no well that's disgraceful I mean I won't allow that sort of thing at all as far as examination papers are concerned it's got to be a capital `q' capital `e' J: Well there's no capital letters it's incorrect Some however did not detect any `spelling mistakes' O: yeah I think that one's Alright, yeah E: Which is the one that's got the spellin' mistake ± elizabeth? I: It's up to you to tell me E: I've got to tell you? ± what if you can't spell?. . . E's comments both here and p. 71 above (`maybe more out of memory') showed how, in his case at least, reading is not dependent on the construction of a word from its component letters, but rather on prior recognition of the word(s) in question 85

What do Lay Speakers Say About Words?

(see also M's comments `I recognise that therefore I know how to say it' p. 79 above). Thus the ability to spell is not a requirement for the ability to read. The aim of this chapter has been to show that linguistic form can be, and is, viewed from different perspectives. A `word's identity for example, was seen from at least four viewpoints ± orthographic, historical, semantic and situational (p. 50ff. above) ± even though there was a prior identi®cation of `words'. When the informants were asked to identify the words in a sentence (p. 55ff. above), a variety of methods were employed including what they thought they had been taught about wordhood, lexicographical practice (pp. 63±4 above), and explicitly drawing a comparison between words and letters, ®gures and/or abbreviations (p. 57ff. above). The same inherently `fuzzy edges' not surprisingly applied to those units below the level of words, commonly known as af®xes, technically as morphemes (p. 64f above). The conclusions drawn in the second part of this investigation are that the relationship between spelling, reading and pronouncing is complex and not simply dependent on a high standard of literacy, that the spelling of certain words causes perceptions not solely related to the dimension of literacy (visual morphemes) and that such morphemes are not `visualised' by all in the same way. Finally, the identity of letters is seen to be just as complex and variable as that of word units, despite the fact that there is a ®nite number of orthographic letters. Linguistic form does not only depend upon recognition of units, but also upon the repetition and quanti®cation of such units both at, above and below what segregationists call the word-level. And such units rarely corresponded to those units called by the professional linguist morphemes or lexemes.

86

4 Words and Linguistic Meaning

The other `plane' of linguistic units ± the meaning ± is just as contentious and variable as `linguistic form'. But this fact is rarely acknowledged by traditional, segregational linguists. For example, John Lyons writes: I will begin by assuming that everyone knows, in a general sort of way at least, what a language is and how it is used. I will also assume that all languages have words and sentences; that both words and sentences are meaningful; that the meaning of a sentence depends, in part, upon the meaning of the words of which it is composed; and that everyone reading this book can identify and interpret the words and sentences of any language, including English, in which he or she is competent (Lyons 1981: 17±18). My informants, instead of merely identifying and interpreting words and sentences of English, came up with meanings, interpretations and identi®cations of words that were not only partially divergent but were also quite ingenious. On the other hand, the alternative view adhered to by some linguists (e.g. Bloom®eld 1935: 139ff, Chomsky 1957: 93f)1 ± that speakers know very little about questions of meaning ± was not supported 1 `In order to give a scienti®cally accurate de®nition of meaning for every form of a language, we should have to have a scienti®cally accurate knowledge of everything in the speakers' world. The actual extent of human knowledge is very small, compared to this' (Bloom®eld 1935: 139). (continued overleaf)

87

Words and Linguistic Meaning

in the present empirical work. Furthermore, what was `known' by the informants was variable and unpredictable. The ®rst part of Question No. 5 investigated whether informants thought that all words had meanings. This question was interpreted in a number of different ways ranging from a straight yes or no to a rejection of the question: I: Do you think that all words must have a meaning? C: Oh this is where we get into nonsense ± can I stop now? [He did, however, go on to answer the second half of this question] Q and W claimed that this question could not be answered out of context: Q: Can't say out of context W: No it depends on the context Some said that some words do not have meanings `in themselves' but rather are used to give the sentence meaning: I: Do you think that all words must have a meaning A: No I: What sort of words don't have a meaning if not all words have a meaning? A: Well most words have meanings ± I think ± em ± but a words like the ± I suppose it could mean one particular object so yes it's got a meaning in that sense I: So you don't or do think that all words have meanings? A: I think that some words have not got a meaning in themselves but they give meanings to sentences like when put into sentences they actually give the sentence a meaning ± but in themselves they haven't got a literal meaning like cat or y'know any any noun or verb ± like swim ± well that's `Part of the dif®culty with the theory of meaning is that `meaning' tends to be used as a catch-all term to include every aspect of language that we know very little about. Insofar as this is correct, we can expect various aspects of this theory to be claimed by other approaches to language in the course of their development' (Chomsky 1957: 103±4n 10).

88

Words and Linguistic Meaning

got a meaning ± something like to hasn't actually got a meaning in itself but it gives a sentence a meaning I: Do you know how it does that? A: No I don't Some denied that all words have meanings because, as N argued, one can be using the `wrong' word: N: . . . many times you come out with a word like that's wrong it sounds right for what you're trying to describe and you try to say one thing and it comes out as another but it still sounds ± Her interview partner thought that such words would nevertheless have a `use': M: Then again it changes it then because the word has got a use but it's not a real word but you have a use for it, because there isn't one in the English language therefore you sort of make up a nonsense word and it's got you thinking hasn't it N: But as long as everyone knows what you're saying I suppose it's a word isn't it B's answer to the question involved introducing the notion of `private words' the meaning of which was known only to the speaker: B: Do I think that all words should have a meaning? I: Or must have a meaning to be a word? B: Em ± I don't know how to answer that one ± really yeah ± I think they ought to but they don't all have a meaning and I use words which don't have a meaning my own private words and things like that ± I use the word [dusaek] ± [dusaek] hasn't got a meaning but its got a meaning to me I: Is it a word? B: No it's not a word no ± it's a word for me that's all Despite the apparent contradictions in the opinions held about `words' and `meanings' here, B seems to view words as having both a private and a public function. 89

Words and Linguistic Meaning

Instead of saying that words `ought to' have meanings as B above, E spoke of one's conviction that they did: E: Well you're led to believe that ± aint ya? The property of `having meaning' for some informants was tied up with a word's utility: I: Why do you think that all words have a meaning? J: Why do I think that all words have a meaning? Well if they didn't have a meaning what would be the use of having words? There were also various interpretations as to what was meant by `meaning'. H, for example, associated the `meaning' of a word with the idea it conveyed: H: Yes I'd say they have a meaning they they can convey some sort of idea And S with what is implied by a word: S: All words should imply something Others contrasted `meaning' with any utterance or noise made by a particular person: I: Right ± do you think that all words have to have a meaning? K: No not really `cause you can talk a load of old claptrap like you do most of the time L: Like you yeah but ± In a similar fashion R claimed that not all words had a meaning because `they can be a sound'. The three informants above were rather inconsistent here, since earlier they had claimed that there was a difference between a word and a sound or noise. F had originally claimed that all words had meanings, but then later she was forced to explain what she thought the question meant: 90

Words and Linguistic Meaning

I: F: I: F:

Is the a word? Yes So do all words have to have a meaning? No ± but I did misunderstand what you said in the ®rst place ± I thought you meant when people talk or what they say, they're meaning something I: So it's not the actual word that has the meaning it's when they talk? F: Yes I guess that's right Function Words Those who claimed that all words did have meanings were asked to state the meaning of the words to, the, it, and so, termed by Lyons `empty word-forms'. Lyons claims that empty word-forms generally have less meaning than full word-forms, and some even, in certain contexts, are devoid of meaning: `Not only do empty word-forms tend to be less meaningful than full wordforms. Their meaning seems to be different from, and more heteregeneous than, that of full word-forms' (Lyons 1981: 48). However, my informants were not in agreement with Lyons in their discussion of these `empty word-forms'. Many, instead of saying that such words could be omitted in certain contexts, claimed that they were either semantically useful or grammatical essential: N: It's [to] making something make sense ± without it ± it's a very useful word really, many sentences wouldn't make any sense at all without it G: . . . I mean I think there are different sorts of meanings and er looking at that as a tree I know what I mean when I say there's a tree er exactly em other words like to it's very dif®cult to de®ne er but nevertheless they are grammatically essential in order to em construct sentences and so they must in a sense have a meaning whatever you mean by meaning and that's the best I can do 91

Words and Linguistic Meaning

I: Okay what about em the word to `t-o' for example ± that has a meaning? J: Yes em I mean to yes I: What does it mean? J: Well it depends how you use it I: I want to eat J: Well it's well using the word to ± you want ± you will eat something ± but if you didn't use the word to you could er ± no hold no wait ± well it describes it describes ± I mean if you left the word I want TO eat if you left to out it would be I want eat ± I: Yeah? _ does it make that much difference to the meaning whether you say I want to eat or I want eat? J: Yes I think it does because ± I: What does it add to the meaning then? J: Er ± time I: Time? J: Yeah I: What time? J: Well at present I: Well I could say I want eat now J: I want ± er no no no you said I want to eat I: Okay if I say I want eat now J: No no no I WANT TO EAT you said, your sentence was I WANT TO EAT I: Okay say I ± J: And I said you could leave the word out to and it would be I want eat I: Mm J: Well when you say I want to eat it usually means I want to eat at the present time but if you said I want eat it could be it could mean any time . . . Some informants did seem to concur with Lyons's view that such words were not indispensable: I: What is the meaning of `t-o' as in to? F: A direction of something 92

Words and Linguistic Meaning

I: F: I: F: I: F:

What about if you say I want to go or I want to eat? If you put it that way it seems quite unnecessary Unnecessary? Yes . . . So it doesn't have a meaning there? It does ± em ± no not when you put it that way no ± it's just ± it's just that it sounds more polite I suppose

D also considered such words were grammatically and semantically essential and added that their omission would render the sentence aesthetically unpleasing: D: The language couldn't do without it so it must have a meaning because if the language could do without it it's almost certain that it would I: But if I say I want eat . . . would you understand what I meant there? D: Yes ± yes I probably would I: So can the language do without it? D: Em in that sense ± in the sense you've just given me it's quite possible ± yes it is ± but at the moment ± er at the moment ± em the omission of a word like to there makes the sentence look disjointed ± unpleasant ± makes it harsh upon the ear . . . they all have their own contribution to make so for example can intensify the idea ± if you took away those words ± perhaps it wouldn't be possible to ± er get across the same kind of subtleties em ± of ideas through the use of language ± maybe that's why they're there Other claims were that such words were `for ®lling in the gaps in the sentence': I: All words don't have to have a meaning? E: No ± it's just for ®lling gaps in a sentence to make it sound better I: What about the word the? M: That ± 93

Words and Linguistic Meaning

N: M: I: N: M:

That isn't as necessary as to I don't think No Has it got a meaning? Not on its own No ± it's just a ®ll-in word for an object, you can't say I sit on chair, you need something to ®ll the gap in ± THE chair N: It's like when you're at infant school you get the cat sat on the mat I mean you didn't say cat on mat would you?

According to these informants above, the utility of such words was not dependent on the meaning (but cf. J. p. 90 above): I: N: M: N:

What about so what Very useful It doesn't make very particularly much sense No it doesn't but ±

K and L claimed that they were `funny' and/or everyday words that are taken for granted: I: L: I: L: I: L: K: I: K: L: K: I: K: L: K:

Does the word `t-o' have a meaning? `T-o'? Yeah To go to the pub ± go to What does to mean there? Are you going TO the place ± TO It's hard to say aint it? ± oh this is a wind up aint it? ± to er I dunno it's a funny word really ± you're going somewhere aint it? . . . What about the? . . . I dunno it's a funny it's funny words aint they? Everyday words aint they? Yeah Does the have a meaning? No I shouldn't think it does really Well you just take it for granted really don't you The cat sat on the mat ± dunno it's just always within sentences innit the ± the 94

Words and Linguistic Meaning

H contrasted `meaning' with `de®nition': I: What's the meaning of the word to as in `t-o'? H: Hmm ± maybe they don't all have de®nitions I guess ± to ± to the store I: Does it have a meaning? H: Er it links up words in the sentence but I don't think I could give you a de®nition for it I: Or I want to eat H: I don't think it has its own separate de®nition . . . I: Does it have a meaning I mean is meaning the same as de®nition? H: I don't see they're much more than just making sense of the words in a sentence em I don't know if it has it's own separate de®nition G also introduced the notion of de®nition but instead of referring to the de®nition of such `form words' as H above, he instead spoke of such words being used to `de®ne something': G: . . . I think that's a bit easier em the de®nitely does de®ne something I think that puts something into some sort of perspective Although G acknowledged the dif®culty with the concept `meaning': G: . . . it rather depends what we mean by meaning doesn't it em which is a very dif®cult question he still wished to claim the `empty word-forms' had meaning (although such a claim was rather inconsistently expressed): I: If I said what is the word to in I want to eat? G: To eat yeah well there I would say it was simply being used as a function word which in a sense no I agree it probably doesn't have a meaning if whatever we mean by meaning you know which is a very dif®cult question em ± hmm that's a very very dif®cult one sorry I can't go very much further 95

Words and Linguistic Meaning

than that but I mean ± yes I would still think it had a meaning but it rather depends what you mean by meaning When a number of contexts were introduced containing the words in question, as so in so much and so what, informants occasionally admitted defeat: G: Em so what? em mm yes you've probably got me there I don't know about . . . E: I aint got a clue Hayley ± why are you making all these little words up from anyway? . . . . . . I never really thought about the English language like this before quite confusing innit ± maybe it's just a saying `cause everyone can understand `em Although J had claimed that the word to had a speci®c meaning in any given sentence (p. 92 above), in an expansion of the same sentence, the meaning of to was now reduced to that of accentuating a verb: I: If I say I want to eat now and I want eat now what difference J: no wait I: does to mean? J: I just think it just, it the word to, it describes what you really want to do I: Eat? J: Yes well using the word to it actually it actually accentuates the word eat, so you know but when you leave out the word to it doesn't to me it doesn't describe really you want to actually to eat, I know you say eat but it doesn't seem to describe the act of eating I: Well what is its meaning in itself ± to there? J: Well it actually accentuates the the doing word, the eating I: So it accentuates another word? J: Yes I: That's its meaning? J: Well that's what I think yes 96

Words and Linguistic Meaning

Describing what such words `meant' often equated to identifying where they occurred in a sentence or showing their relation to other words: I: And the has it got a meaning? P: Well it's got a meaning in a sense, it has got a meaning, it's like at the start of the sentence or something H: The ± I guess you usually use the in front of a noun showing that it's singular I guess O: . . . that could mean anything, it's the most commonly used word in the language I would have thought I: It's got lots of meanings? O: Yeah I: Can you tell me the meaning of to? . . . G: . . . it means a vast amount of things ± em it's used before a verb I: E: I: E:

What about the word the ± what does that mean? The beginning The beginning? The beginning ± it's always the beginning of a sentence innit?

D: Er to is used there to create er ± the in®nitive of the verb I: But what does it actually mean? D: I think it has to be taken ± I think it has to be taken with the verb B claimed that in such cases the verb `has the meaning' I: So what's the meaning when it's used before a verb? B: Er yeah ± I don't know I: But you still think it's got a meaning when it's used before a verb? B: No ± no it hasn't got a meaning when it's used before a verb ± the verb has got the meaning ± to my way of thinking anyway 97

Words and Linguistic Meaning

On the other hand, V denied that to had a meaning `because on its own it doesn't mean anything'. Such lay metalinguistic judgements also cast doubt on the widely held opinion (Atkinson et al 1986: 191 ff.) that function words, although they may create dif®culties of precise semantic identi®cation, nevertheless do make a distinctive semantic contribution to the expressions in which they occur. Atkinson et al cite as proof for their axiom that there `are no words in a language which lack meaning' (ibid.: 191) the `Principle of Compositionality' (see p. 100 ff. below for a discussion of their principle.) But there is some evidence that a number of informants (e.g. E: p. 96 above `maybe it's just a saying') may have been treating the grammatical word-forms in a speci®c phrase (e.g. so what) as `lexicalized' forms (cliches/idioms etc.), i.e. units displaying a `syntagmatic interdependence' (Saussure 1922: 176). The Dictionary Very often, the ®nal court of appeal in ascertaining a word's meaning was the dictionary: G: Well if I had a hundred years I could tell you ± look through the Oxford English Dictionary it means a vast amount of things H: Well I'm sure all these words mean something behind them in the dictionary but but I don't know for sure what they would be so I don't know what its de®nition would be B: Well if you look up to in the dictionary I think if I remember rightly it it describes it in the way that I did describe it like when it's used as in the way of `towards' like We're going to Paris like we're going to Paris to you know Thus, as might be expected, the dictionary played a key role in informants' understanding of what is meant and understood by the `word'. Lexicographical entries were seen as arbiters of disputes concerning the spelling and meaning of words. In addition, some informants appealed to lexicographical omission 98

Words and Linguistic Meaning

as a justi®cation of `non-wordhood' (e.g. O and A, pp. 63±64 above). Others, as noted above, invoked lexicographical de®nition as the criterion for a word's meaning. However, a direct questioning of informants on dictionary de®nitions as in Question No. 7, produced answers which often contradicted the above points of view. Although, for A, a criterion for being a meaningful word was its inclusion in a dictionary, he was critical of all the de®nitions cited from the COD: A: . . . it's far too short ± it doesn't tell me what a cat is at all ± that could describe a rat because you could have a domesticated rat [cat] I thought it was meant to relax you [yoga] not a clue wouldn't help me in the least [ontology] That is hogwash. The only word in it that gives you any idea of what it is is the word insect [ant] B was also not totally in agreement with these dictionary de®nitions B: . . . don't think that's totally right I: Why not? B: Because a cat ± certainly furry ± certainly not always domesticated . . . I: So you think the dictionary there is wrong? B: They can be but not always `cause lets face it a lion and a tiger and all that lot there are cats and I'm not going to call a lion and a tiger small and domesticated, for Christ's sake ± tell the stupid bastard who wrote that out to go walk into a lion's den sometime especially when she's got em cubs Other informants criticised the de®nitions, not for being inaccurate, but for being too complicated and confusing: P: I just thought it was a form of exercise, I suppose that's what it's supposed to be innit, don't really describe it well though it sort of describes it but it could be put much simpler couldn't it [ant] 99

Words and Linguistic Meaning

N: It's all very well putting that in a dictionary but how could anybody read that and know what hymenopterous is? If they're looking up ant they're hardly gonna know what hymenopterous is ± I think that would make someone more confused than when they started if they looked up ant E: it's the posh way of sayin' it innit . . . they've got half of it right aint they then I don't see why they have to put these dirty great big words in it ± everyone knows what an ant is don't they According to the lexicographer Sidney Landau, criticisms such as those voiced above, although common, are misguided and misinformed, confusing ostensive de®nition with meaning. Moreover, he claims, the employment of such `technical' or `hard words' having `more exact meanings' does not require any further quali®cations and therefore saves lexicographical space (Landau 1989: 136±7). But as Chapter 6 below shows, the criteria for a `hard', or for that matter `simple' word are far from obvious. Thus although my informants generally used lexicographical inclusion or omission as a criterion of wordhood, they tended to be rather critical of the de®nitions themselves. Moreover, both H and E (p. 96 and 98 above) seemed somewhat confused as to the purpose of lexicographical practice. Ricks hints at this rather paradoxical aspect of the dictionary: For the meaning of the word is not a matter of fact (which is why an argument about it can't be settled by recourse to the dictionary), and it is not a matter of opinion (which is why an argument about it mustn't be unsettled by a refusal to have recourse to the dictionary). The meaning of a word is a human agreement, created within society but incapable of having meaning except to and through individuals (Ricks 1980: xi). The Principle of Compositionality The view that some words do not have meaning, or that they do not have meaning on their own but are simply employed to 100

Words and Linguistic Meaning

highlight the meaning of other words, calls into question the commonly-held assumption that understanding and producing sentences is a compositional process. The principle of `compositionality', that `the meaning of a composite expression is a function of the meanings of its component expressions', has, according to Lyons, a philosophical tradition which was ®rst given expression within linguistic theory by Katz and Fodor (Lyons 1987: 164).2 As Baker and Hacker argue, this principle cannot provide any guarantee for excluding `nonsense' (since there are `no tolerably general principles which lay down necessary and suf®cient conditions for combinations of words to make sense' (Baker and Hacker 1984: 336)) nor can it, being solely concerned with `circumstance-invariant features of type-sentences', make sense of many sentences which would immediately be understood in context. The utterance `give the coffee to the spotted dick', makes perfectly good sense if one knows that it is an instruction from one waiter to another. And when my son, on ®rst seeing an open-top bus, made the comment `oh look, there's a bus without a lid on', I understood perfectly what he was drawing my attention to. As Baker and Hacker conclude: If a compositional theory grants sense to any combination of words which might conceivably be used to say something intelligible, much of what passes for nonsense must be reckoned among expressions which really make sense, and it will be dif®cult de®nitely to identify any outright nonsense at all. If, on the other hand, nonsensical expressions are admitted sometimes to make intelligible statements, the theory must abandon the thesis that what a 2 `In encountering a novel sentence, the speaker is not encountering novel elements but only a novel combination of familiar elements. Since the set of sentences is in®nite and each sentence is a different concatenation of morphemes, the fact that a speaker can understand any sentence must mean that the way he understands sentences he has never previously encountered is compositional: on the basis of his knowledge of the grammatical properties and the meanings of the morphemes of the language, the rules the speaker knows enable him to determine the meaning of a novel sentence in terms of the manner in which the parts of the sentence are composed to form the whole' (Katz and Fodor 1963: 482).

101

Words and Linguistic Meaning

speaker says must be calculated from the sense of the sentence uttered. Compositional theories of meaning are stranded in a no-man's land, caught in a deadly cross®re (Baker and Hacker 1984: 338). Some of the responses to the third question of Question No. 3, showed that familiarity with, and understanding of individual words was neither a necessary nor a suf®cient condition for comprehending a phrase or sentence. Informants were asked to comment on the de®nition of ontology. Even though the majority of the subjects knew the words nature and being, many had dif®culty with the phrase nature of being I: N: I: N: M:

What do you think ontology is then? Studying the nature of being whatever the nature of being is What do you think the nature of being is? That's what I'm trying to think of, it's dif®cult to describe that It's like why are plants so hard to grow, WHY well why I suppose N: I don't honestly know ± I mean the nature of being it sounds ± you can understand it but you can't explain it what the actual nature of being is M: The only way you can really describe the nature of being is nature of being Informant H, when pressed to analyse his dif®culty, singled out the word being as creating a problem in this particular context. I: Okay do you know the meaning of ontology? H: No I don't [reading] that branch of metaphysics dealing with the nature of being I: Do you know the meaning of ontology now? H: No . . . . . . I: Why? H: Em ± well now nature of being is that as in dealing with why we exist? I: You're having problems with the phrase nature of being? H: Yes 102

Words and Linguistic Meaning

I: H: I: H: I: H: I: H:

You know the words nature and being? Yeah So why don't you understand it? [laughs] the nature of being well being would er okay probably being is my problem word You don't know what ± Nature of being is that just ± if I was guessing nature of being I'd just say it was how we exist and why we're here What is the dif®culty with the phrase nature of being or with the word being? ± you know what the word being is? It's hard to put into words but yeah ± am I not right in what I think it might be?

O guessed `I suppose it means being at peace with yourself, I wouldn't really know', whereas J thought it meant `facing reality'. Hacker elucidates the foundational misconception stemming from compositional conceptions of meaning: The misconceptions ¯ow, inter alia, from the idea that there are, in a language, sharply de®ned category rules which apply in all contexts. But while I know what `I feel water with my hand' means, and what `a foot underground' means, and do indeed understand the sentence `I feel water 2ft underground with my hand' uttered by someone poking his arm down a hole, I do not know what `I feel water 30 feet underground with my hand' means, when said by someone in similar circumstances. It can be given a meaning easily enough, but it does not have one merely in virtue of its composition out of meaningful constituents (cf. BB, pp. 9f) (Hacker 1986: 185). Such criticisms made of the `principle of compositionality' can also be made of `orthodox' treatments of metaphor which generally interpret metaphorical understanding on the basis of a prior knowledge of the literal meaning of each individual word (see p. 107 below). 103

Words and Linguistic Meaning

Concept Words Wittgenstein claimed that many of the erroneous views concerning the meaningfulness of words arise from a lack of `surview'3 over our ordinary use of language. We are beguiled into thinking that the facts of language must conform to `pictures' evoked by certain uses of language: what confuses us is the uniform appearance of words when we hear them spoken or meet them in script and print. For their application is not presented to us so clearly. Especially when we are doing philosophy! (Wittgenstein 1958: 11). Although we may think that words such as to function analogously to the word chair and number, such conceptions are misleading: we can give an `ostensive'4 de®nition of the meaning of chair and we can give a grammatical description of `empty word-forms', but we can not explain number in the same way. We can show the meaning of number by giving examples of particular numbers, yet this method of explanation is considered inferior because of the `pictures' we see in our uses of language: Our concept of a substantive expression is, as its name suggests, constructed on the paradigm of a name of a substance. Consequently it is a natural inclination in us to look for a substance for every substantive. We exacerbate this inclination when we couch our philosophical, conceptual questions in the form of questions about entities, e.g. `What is length? What is meaning? What is number?' instead of `what is an explanation of meaning? How are lengths measured? How are numerical expressions used'? Consequently we assume Platonistically the existence of special objects to correspond to substantival expressions for 3 `Surview' or `synoptic view' is used by Hacker (1986) as a translation of the È bersicht (from U È bersehen, roughly meaning `to command a clear German U view'. 4 Ostensive de®nition is indicating an object (usually by pointing) and uttering its name (see Wittgenstein 1958).

104

Words and Linguistic Meaning

which we can ®nd no ordinary objects. We talk of numbers as ideal objects, obscuring the fact that all that can coherently be meant thereby is that the use of numerical expressions is in certain respects similar to that of signs that stand for objects even though numerals do not (RFM, pp. 262f) (Hacker 1986: 169). When my informants were asked to state the meaning of number, all gave examples of numbers as an explanation of the meaning, yet some did not appear happy with this form of explanation: H: I would probably say number has ± a speci®c meaning I: Such as? H: Well to number something off like a verb to actually count `em down or number ± a number is each of the little ®gures in our numbering set ± I don't know zero through nine they're they're each a number em what else? G thought that I was `teasing' when I asked him to give the meaning of number without using examples: G: The word number well to me it means em something like ®ve or six or seven I: Can you tell me what the meaning is without giving examples can you de®ne the word number? G: Em ± I'm not sure that I can you're teasing I'm not going to go on with this [laughs] it has reference to those things which we call ®ve or six or seven whatever they might be D explained the word by reference to the objects or ideas under consideration: I: D: I: D:

What about the word number? ± does that have a meaning? Er yes ± yes it does What does it mean? It means em ± number means the sum of ± er ± any number ± the sum of any er ± collection of individual objects or ideas that's being considered 105

Words and Linguistic Meaning

I: What about when you say one two three go? D: Em ± yes ± those are numbers ± the ideas that were being referred to I would have thought were seconds ± therefore you're talking about a number of seconds ± in this case three For M, number, as well as having a speci®c meaning, referred to drawn shapes: I: And what's the meaning of the word number? M: I you say there were a great number of people ± N: You could say lots you wouldn't have to say number do you? And when you say number you obviously think of ®gures don't you? M: It's when you draw ®ves and twos and things they're not actual letters they're ± it's it's what you actually call the shapes you draw I: What does it mean? M: It means ± amounts of things O offered the widest explanation of number O: Well it can mean umpteen many things it can mean a song it can mean a number of a house basically what it is number, to me if you say number I would immediately think of a number of a house, you know it's a number like a two or a three or a four ± not necessarily a song As is seen here, our concepts are not sharply de®ned, and, as Wittgenstein wrote, `Giving examples is not an indirect means of explaining ± indefault of a better' (PI, 71). Rather `Examples are decent signs, not rubbish or hocus-pocus' (PG, p. 273). In the same way, pointing to a pair of bananas is a perfectly adequate explanation of the number two. As Hacker writes, conceptual investigations therefore `are investigations into our measuring rods and their uses, not into what is measured. Conceptual problems are toto mundo distinct from factual, scienti®c ones, and cannot be resolved by scienti®c advances, 106

Words and Linguistic Meaning

but only by clari®cation of the use of words (Hacker 1986: 196).5 Metaphor The observations above, that language-games are heterogeneous, and thus the meaning of a word is dependent not only on the context, but also on an individual's interpretation of the context, call into question the notion that a metaphor `consists in giving the thing a name that belongs to something else' (Aristotle Poetics: 21), that it is a `®gure of speech in which a name or descriptive term is transferred to some object different from, but analogous to, that to which it is properly applied' (OED2). Such conceptions again involve questions of identity and similarity: to speak of the transferring of a name from one word to another would seem to imply that words had some invariable `portable essence' independent of a particular understanding of them by a speaker. To reject the above assumptions may appear to be denying the concept of metaphor, a concept employed by lay people. A metaphor, for many people, seems to apply to sentences which one could not imagine being used literally, sentences such as `A red apple is a snivelling dormouse'. Certain sentences of Question No. 4, viz questions 1, 2, and 3, investigated whether, and if so how, the lay person construed the literal/metaphorical distinction. Some informants only assigned an interpretation to these sentences and did not draw on the notion of literal meaning at all (O and E, p. 108 below), others commented on the perlocutionary6 effect of using such utterances (F and P p, 108 and 109 below) and others on the `correctness' of using such expressions (W and J, p. 112 and 113 below): 5 One can here be critical of Bloom®eld's statement that `We can de®ne the meaning of a speech-form accurately when this meaning has to do with some matter of which we possess scienti®c knowledge' (Bloom®eld 1935: 139). 6 This term is from Austin 1962, and refers to the effects or consequences that speech acts (utterances) have on the beliefs, thoughts or actions of the hearers.

107

Words and Linguistic Meaning

O: Exactly what it means John is a pig he's an animal er ± John is ± I: Actually he's an animal? O: John is an animal yeah, John is a pig presumably the man eats spaghetti bolognaise like I do everyone gets a piece of it but I think that's what that means anyway ± are those glasses dead? Are those glasses dead? Empty glasses in a pub. I: Why d'you call them dead? O: Well they're empty they've got to be washed up before they can be used again ± in most good hostelries ± in most good hostelries ± he was mad at me well girlfriend stays out late ± the bloke gets mad at her I: F: I: F: I: F: I: F:

How is the word pig used there? Vindictively Anything else? What in that sentence? Yeah No ± right are those glasses dead? that's asking questions How's dead used there? Well as as in as in asking the question ± dead ± still quiet no not quiet I: Are they both used in the same way pig and dead? F: No no not at all pig's vindictively and dead's just just sentences just as far as questions are ± he was mad at me well that's just em ± that she was saying that she's upset and he's obviously angry

H: That's just saying John is messy he needs a clean up . . . em he was mad at me mad me being upset or angry E was even pushed, unsuccessfully, to acknowledge some kind of distinction: I: E: I: E:

Just comment on it ± on the word pig there ± John is a pig John is a pig ± that's an insult Yeah what does it mean? That he's a pig 108

Words and Linguistic Meaning

I: E: I: E: I: E: I: E: I: E:

That he's a pig? Yeah Okay ± are those glasses dead? Means they aint got no life in `em ± beer's been drunk Is it used in the same way as pig? No Why? 'Cause it aint So what does dead mean there? Empty glasses

P: Pig's sort of like insulting somebody a bit ± are those glasses dead? is just like a slang word for saying somebody ®nished them it's just an expression that people use isn't it? For the ®rst two sentences, N and M only explained the meaning: M: Em pigs are supposed to be dirty smelly greedy N: Yeah greedy I mean somebody who's who ± I always call Mick [her boyfriend] a pig M: It's supposed to be unpleasant N: Yeah it's an insult de®nitely M: Are those glasses dead? N: `Are they ®nished with?' M: Yeah N: How is the word dead used there? M: Dead's `®nished' N: Dead's you're not drinking out of them anymore However, when questioned about mad, these informants then appealed to the `literal' meaning and presumed country of origin of such an expression: M: He was mad at me ± sort of angry at me N: Cross angry yes I: Is that what the word mad means? N: Not literally no but you ± it's sort of an American way of putting it 109

Words and Linguistic Meaning

M:

It's a mental disturbance ± a mental disturbance I suppose to a degree if you're really angry you get to a degree where you're not thinking straight as you usually would N: But that's not necessarily what that means is it? ± that's just an American way of saying he was angry

Q also said of the sentence above that it was an American usage: `'real mad' means angry'; he did not offer any further comments on this use. Other informants, although they were able to paraphrase the sentences under investigation, nevertheless said that they were `impossible'. A claimed that pig referring to a person was not used literally, dead applied to `glasses' was impossible but meaningful, but mad, in the above context, was used in its `other' meaning: A: John is a pig ± pig being the underlined word ± John could not be a pig literally ± that would be impossible and pig is used as an expression to say that he's not a very nice person . . . are those glasses dead? It's impossible for the glasses to be dead because they weren't alive in the ®rst place but it means that the glasses are empty I: So how is the word there used? A: It's a slang expression used in bar terminology I: Is pig not slang? A: Yes I'd say it was slang I: So those two words are used in exactly the same way? A: Yes ± yes ± he was mad at me ± em mad is used to say that someone is angry but it can mean that I believe ± it has another meaning the word mad but it's not used in that context ± it's to describe that someone was angry at me B offered an evaluation of such sentences: after stating what these sentences `meant', he claimed that they were `silly' and `stupid': B: Well it's an insult ± em ± implies he isn't a very pleasant person ± could imply he eats a lot ± it's very unfair on the animal they're rather nice things ± nothing like as dirty as 110

Words and Linguistic Meaning

I: B: I: B: I: B:

people make them out to be ± are those glasses dead? ± that's a really funny one I always laugh when I say it myself ± at myself ± implies sort of you know glasses lying around with bullet holes in them or something ± em meaning em ± have they been ®nished? . . . Can you see any similarities or differences between John is a pig and are those glasses dead? Yeah they're both silly They're both silly? Yeah Why? Well because as I say how can you call a human being a pig for a start because a pig is a ± as I said it's an animal and it's being remarkably unkind to the animal because you're implying that the pig's an absolute shithouse and in fact the pig isn't and are those glasses dead again it's stupid isn't it

Only one informant, C, introduced the term metaphor when speaking of these sentences and, in so doing, differentiated between various types of metaphor. Moreover, he was able to assign both a literal and a metaphorical interpretation to such sentences: C: Well actually it's a complementary noun or uncomplimentary [laughs] noun er ± it well that's what it is ± it means `John is like a pig' it means it's not literal I: So how is the word used there? C: Well metaphorically I: What is a metaphor? C: Something that isn't literally true I: Do you think it's always possible to tell whether something is literally true or metaphorically? C: No I don't think it is always possible ± I mean that er ± that is used in a pejorative sense but em ± you could be pointing to your pet pig and saying John is a pig I mean if you were pointing to it it would be stupid to point it out unless the blperson was blind then you wouldn't be pointing to it ± are those glasses dead? ± good one ± em . . . . I: . . . Is it used as metaphorically as in the ®rst case or not 111

Words and Linguistic Meaning

C: No it's a bit different it's so specialized that I wouldn't say it was a metaphor in the ordinary sense em ± I mean obviously it's not literally true because your glasses don't live or die I: He was mad at me C: Well mad there is just a it's a kind of abbreviation for mad with anger so again again it's you know not a metaphor ± I suppose it's just a kind of specialized shortening of a phrase if someone is very angry em ± you know you refer to people losing their head with anger or going mad with anger I mean it's just a shortening of that ± doesn't mean he's insane P also, like C above, but without using the terms metaphor or literal meaning, distinguished between the sentence being used to refer to a person or a real pig: P: John is a pig it's just like to express somebody innit depending on whether John is a person or John is a real pig I: What about if he's a person? P: If he's a person? It's like running him down innit Although Q did not use the word `metaphor', he spoke of pig being `used in an extended way . . . the meaning has been stretched' echoing a common view of metaphor, that it `is held to be an extended use of a word' (Taylor 1970: 165±166). However, his opinion on dead was similar to C's above: Q: It's not strange ± it has a special sense ± it's a set phrase a pub word ± you can't separate it from the phrase W however commented merely on the `correctness' or grammaticality of such expressions. Referring to pig he said W: It's an incorrect expression ± it's common but you know what it means And to dead W: It's a correct expression as long as you don't want to win prizes for grammar 112

Words and Linguistic Meaning

whereas he considered mad to be an `improper use of mad'. J also spoke of the `incorrectness' of such sentences: I: J: I: J: I: J: I: J: I: J: I: J:

I: J:

So you're saying it's a bit odd to say John is a pig? Yes because you're not saying a man is a an animal But he's not? He's not So do you think it's wrongly used? Yes . . . . . . He was mad at me ± well again it it's not correct Why isn't it correct? Well there's mad ± mad means he was insane ± you can't have ± he he can't be insane at me Do you think that sentence is ever used? They do use it yeah And do they mean by it he was insane at me? No it means he was annoyed . . . . . . they can't be dead but the problem is it's used in a public house or something ± where they're not emptied [the beer glasses] ± it's a phrase they use Is it incorrect? Well it's incorrect but the people ± but the publicans or people working in the licensed trade would understand because it's it's a phrase er appertaining to that sort of trade

D, in contrast, claimed that pig made sense because of the similarity between the animal and the man but dead in the context of glasses did not: mad he said, was being used in a specialized sense: I: Right can you comment on how the underlined words are used in these sentences ± John is a pig D: Abusively ± er ± in the ®rst one ± er ± second ± em it doesn't ± it doesn't make sense but perhaps it's used as a joke I: Why doesn't it make sense? D: It doesn't make sense because the verb dead em ± can't refer to inanimate object which don't possess any life I: But pig can refer to humans? 113

Words and Linguistic Meaning

D: I: D: I: D:

I: D:

I: D:

Yes Why can't dead refer to glasses? Well dead is ± er ± dead's a verb there A verb? Yeah ± is that right? ± they are dead ± yes ± no no it's not a verb it's actually an adjective ± there ± em ± has inherently ± em ± an idea which is not ± er ± relevant ± not shared by the inanimate glasses because the idea of death involves the absence of life ± em ± in objects which have lived or do live whereas glasses don't live at all ± they have no life Well why can pig be referred to John? Em ± well because there's a certain kind of plausibility in this sense in that both animals are alive ± man in the case of John and pig ± they share that ± perhaps that's where the plausibility in a sense comes from ± but in addition to that there is the idea that pig in some way is a lower form of animal and that's why such an effective term of abuse is there Mad Yeah ± mad there is used in a very kind of specialized sense ± em it doesn't have its conventional meaning madness that is of insane ± er ± instead in that context it has the meaning of ± er ± extreme annoyance expressed by someone

Informant T was unique in not being able to make any comments on such sentences. When asked how these sentences were used, her reply for all three was `don't know': one did not know whether her inability to pass any judgement was necessitated by her, for example, lack of understanding of what she was supposed to be commenting on or by an incomprehension of the sentences themselves. The variation within such comments on the sentences under investigation, that for example, `they make sense' but are `impossible' or that they are `just expressions' or `not literal', `silly', `incorrect' or `common', shows that metalinguistic, metaphorical games are of the same kind as all questions concerning word meaning(s), such as sebum means `the semi-liquid, greasy secretion of the sebaceous gland', arbre means `tree' in French or wicked means `good' in youth culture. In all 114

Words and Linguistic Meaning

such `games' participants are being asked to show their understanding, by means of a reformulation, of the relevance of the words used in a given communicational situation. Moreover, the interpretation of these sentences does not depend on the concept of `literal' or `original' meaning, pace Grice as formulated by Lyons: The speaker/writer cannot mean that literally [John is a tiger] . . . He must believe (if he is being co-operative) that I can work out the non-literal meaning for myself ± presumably on the basis of the literal meaning (of the whole utterance-inscription or of one or more of its constituent expressions) (Lyons 1981: 215). But if speakers can readily make sense of `metaphorical' expressions, then it would seem that they are obviously not interpreting such expressions on the basis of the `literal' meanings of the individual words. Moreover, some informants did not appear to be treating the word pig in the sentence John is a pig any differently from its use in referring to the farmyard animal. If this is the case, the understanding of metaphors is, no less than the understanding of individual words, culture- and context-dependent, as is especially highlighted in the sentence involving dead glasses. One aim of this chapter has been to indicate the problems with semantic identity within orthodox linguistic theories and explanations and, in so doing, to highlight those features of language that give rise to the greatest confusion (see pp. 103±106 above). Despite one view that `normal' native speakers, in one sense, `know all there is to know about language'7, the fact that such language-users generally lack a surview (p. 104 above) over everyday linguistic practices is illustrated by the informant responses. A common view, seen both in this empirical work and in technical linguistic theory, is that words qua words, have an identical function: viz that of having a representational relationship to other entities ± either to ideas (H, p. 90, D, p. 93 above), 7 See Davis (1997), Hacker (1986: 152±3), Harris (1980 Chapter 1) and Love (1998: quoted pp. 33±34 above) for an elucidation of this point.

115

Words and Linguistic Meaning

or things (A, pp. 88±89 above). It is this relationship which, for some informants, explains lexical meaningfulness (J, p. 90 above). Hence, any discrepancy between words and what they name is in some way faulty (N, p. 89; K, and L, p. 90 above). This chapter has also illustrated that (a) identity of meaning or interpretation, is not a necessary condition for successful communication and that (b) any attempts at securing an `identical uptake' would be a fruitless endeavour. For example, the informants had different interpretations when presented with the apparently `straightforward' question `do all words have a meaning?' B, (p. 89) appeared to interpret this as referring to codi®ed or institutionalized `meaning', whereas H, (p. 90) to the expression of an idea, and E, F, G, H and P 91ff. above), when asked about `form words', to the position in which a word appeared in a sentence. The question `how are the underlined words used?', showed the various ways in which the informants interpreted the point of my question. B, (p. 110f. above), may have thought that he was being asked for a subjective evaluation, W and J (pp. 112±113) a comment on the dimension of `correctness' and A (p. 110) seemed to appeal to a codi®ed authority (`It can mean that I believe'). Finally, it is signi®cant that two informants, D (p. 113) and T (p. 114) either failed to understand the sentences or the point of the question since, in both such instances, the words in question were common everyday words. Although it may be readily accepted that speakers in interaction are constantly negotiating meanings, some still may object to the `indeterminacy of meaning' thesis on the grounds that, from many contexts, it is possible to extract or abstract general or speci®c meanings which, for example, the language-learner may take as guides, perhaps even `cores' of meaning as ®rst hypotheses in communication. And dictionaries and grammars function precisely in this respect. But this is to approach the issues the wrong way round: the dictionary, and other such institutions, are responsible for this lack of surview identi®ed by Wittgenstein and by Baker and Hacker. By collating an `average' or `representative'sample of words in a text or in a sample of texts, the dictionary has de®ned what people in the past have meant by a given word. There is no guarantee that speakers will mean that in the future. 116

5 Parts of Speech and Grammar

The traditional terminology for the groups of words that are generally called `noun', `verb' etc. was `parts of speech'. Modern linguists have on the whole rejected this label, preferring the term `word classes'. The traditional terminology is accused of employing unsatisfactory de®nitions: `the de®nitions are largely notional and extremely vague' (Palmer 1984: 55). Palmer goes on the attack against such notional de®nitions, claiming that de®ning a noun in terms of naming things `is clearly a notional de®nition at its worst' since it presupposes an unambiguous application of the term `thing' to all nouns: `how do we know what a thing is? Is ®re a thing? Is peace? Is hope or intention?' (ibid.: 39). The preferred modern analysis is generally on distributional grounds such as those laid out by Zellig Harris in 1951 ± i.e. `items or strings of items that are mutually substitutable and can occur in the same set of environments are assigned to the same category' (Brown 1984: 93). However, when a sentence is analysed in this way, some linguists discovered that the traditional and modern analysis coincided in all but the de®nitions and description of the classes discovered: `Provided we are aware of the problems we can use the traditional parts of speech and their terminology as the basis for word classi®cation' (Palmer 1984: 58). This was similar to an opinion expressed by Chomsky twenty or so years earlier when he wrote that the information presented by a traditional grammar is, without question, substantially correct and is essential to any account of how the language is used or acquired' (1965: 64). And O'Grady, when discussing how to 117

Parts of Speech and Grammar

determine a word's category, prefers to sit on the fence saying that meaning, in¯ection and distribution can all be helpful. However, for him distribution is often the more reliable criterion (1996: 183±4). However, it doesn't take much ingenuity to see the ¯awed logic in adopting a distributional analysis. Harris well illustrates this with the word why in John wanted to learn why, the morphological and syntactic evidence for determining whether why is a noun is inconclusive. Here why occurs in a position which can in principle be occupied by nouns or noun phrases . . . but also by adverbs . . . On the other hand, why does not usually combine with articles, possessives or demonstratives, but may occasionally do so (Harris 1981: 66). In addition, the so-called vague, notional de®nitions, were never intended to be criteria for classifying, without exception, grammatical forms. It is hardly surprising to see the layperson's confusions when asked to speak of grammar and parts of speech. As seen in the past two chapters, individuals are able to make comments on the `formal' identity of words ± counting them either as `types' and/or as `tokens' (see pp. 51±54) ± to say whether or not they are `meaningful', how they are pronounced, and whether they are spelt correctly etc. This chapter will look at individuals' understanding of grammar and grammatical aspects of words along with their opinions on how such aspects may be linked up with certain `classes' of people and or different types of situations. These latter opinions, unlike the former, are not a direct consequence of the aspect of literacy which involves learning to read and write; one can be taught the spelling of words without knowing their `part of speech' or the situation(s) in which it is appropriate to use them. Moreover, differing from the comments made earlier on `linguistic form' and `meaning', some informants were more ready to volunteer the origin of their opinions: for example J: `My English teacher told me so a million years ago' [why aint is incorrect]; B: `It's what I was taught' [when asked what `proper English' was]; H: `That was 118

Parts of Speech and Grammar

what we were always told in elementary school, person, place or thing' [the de®nition of a noun]. It is thus indisputable that what and how children are taught can remain (often imperfectly or incompletely recalled) unquestioned well into adulthood. This is also supported by Hamilton and Barton's empirical investigation on adults' de®nitions of the word `Many re¯exes of schooling were observed in people's descriptions of language' (Hamilton and Barton 1983: 590). In fact, they observe that schooling is as great an in¯uence on the understanding of the word as literacy itself: `the relationship between the de®nition of word and literacy level is much less striking than we might expect from the literature'. It is not that literacy effects `a qualitatively different understanding of the concept of word' but rather it helps contribute to `a fuller integration of the concept into all aspects of knowledge of the language system and a more sophisticated expression of it in terms of the de®nitional form used' (Hamilton and Barton 1983: 589). In Question No. 6 I investigated informants' understanding of such grammatical classes as `noun', `verb', `adjective' etc. to see whether they used `distributional' or `notional' criteria in identifying such forms. I only mentioned the terms that had previously been mentioned by the informants themselves; subjects were merely asked to comment on the underlined words. Many volunteered such information in answering other questions. For example, when asked how pig was used in the sentence John is a pig (Question No. 4), the ®rst comment from B was `descriptive noun' and, talking of the identity of words in Question No. 2, C, D, and Q employed arguments pertaining to word classes (see Chapter 3 p. 52. above). A independently brought up the question of nouns and verbs when referring to gazooks in Question No. 3, referring to formal and semantic characteristics of the word class and assigning `plurality' to nouns and sentences but not to verbs: A: Well gazooks sort of em ± leads you to believe that it's a noun ± that it ± em ± that it ± you know ± it it's ± like a book or em a coathook or something like that ± em I: What is a noun? 119

Parts of Speech and Grammar

A: Well it's an object ± you know ± what's the de®nition ± person place or thing or something like that and snortle would tell you to lead you to believe that you were actually ± that it was a verb you were actually doing something ± you know ± like smoking I: Why's that? Why do you say that about those two words? A: Well if you were aliken it to ± if you were to say snortle was a bit like hurdle ± no that could be either couldn't it ± em ± gazooks has got an `s' on the end of it so that would tend to lead you to believe it was a plural of gazook and you couldn't have a plural verb ± and snortle I suppose you could have apple ± that could have an `l-e' on the end of it ± you can have a snortle ± em it could be either I: Why can't you ± why do you say you can't have a plural verb? ± why can't you have a plural verb? A: Well you can only do one action at once can't you I: What about if you're talking about a lot of people who do the same action? A: Those people all swam or whatever then ± er ± I don't think it's a plural verb ± is it? I don't know I: What about the verb were? A: I suppose ± that's a bit abstract ± that's ± I: Is that singular? A: No it it ± er it applies to a plural sentence ± they were or yeah ± that's a good point. Although many informants were able to speak of nouns , verbs and adjectives etc., there was very little agreement as to how to apply such labels. B, G, H and M said that the instances of watch and chair in the pairs of sentences (Question No. 6) were respectively a noun and a verb, but A denied that watch in the second sentence was a verb, since for him, a verb was something used `literally': A: Watch ± in ± used the ®rst sentence is a ± a noun i.e. a wrist watch and watch ± your language is you're telling someone to be ± less ± er ± telling someone to swear less I: But what's ± what's the actual difference between the two words? You don't have to tell me what the sentence means 120

Parts of Speech and Grammar

A: I: A: I: A:

No em You said that was a noun Yes What's the second one? Em it's an order ± innit -it's em ± you're givin' someone an order . . . . . . You're not literally telling someone to watch their language ± because they can't I: So that's not a verb [watch] but that is? [chair] A: Yes

He also thought that love in I love dancing was possibly an adjective: A: Love in the ®rst ± is er ± describes ± how you feel about dancing I: So is it a noun verb or adjective? A: I suppose it would be an adjective in that sense Some informants had a vague or incomplete recollection of grammatical terminology. For all the questions in No. 6, E, P and O only referred to the meanings of the sentences (see pp. 121±122 below): but when asked what the letters a e i o u were called (Question No.9), E said `nouns or verbs'? E: I: E: I: E:

What then ± this is like going back to school Vowels ± Do you know what the other ones are called? Nouns? No consonants Consonants ± yeah ± I watched that programme once ± I should know this.

Here E was probably referring to the popular day-time television show called Countdown in which the contestants are required to choose a number of unseen consonants and vowels. The winner is the person who makes the longest word from the chosen letters. O also remembered some of the terms but he did not know what they referred to or what grammatically meant. 121

Parts of Speech and Grammar

I: Do you know grammatically the difference between those two words? Can you remember? O: I have a watch and watch your language well one's an article a wrist watch and watch your language is someone telling you how to behave really ± is that right? Grammatically I can't remember what grammatically means, I was good at English at school I: Do you remember nouns and verbs? O: Adjective is a describing word, that's about all I can remember, I can remember the names but I can't remember what they mean. P did not mention the `parts of speech' at all, even when asked whether she remembered, from her schooldays, the types or classes of words: I: P: I: P: I: P: I: P: I: P: I: P:

Do you know the names for the words watch and chair? What do you mean? What were you taught at school what they were? Which that watch? Yeah That's the watch on your arm What's the type of word, the class of word? What they're actually an object d'you mean? Yeah, what words ± Oh I can't remember You can't remember? No I can't remember I'm afraid

When asked to de®ne such classes as `noun' and `verb' many admitted defeat, especially after some such `problematic' words as love and hatred were introduced (Question No. 6, question 3). B was unable to satisfactorily answer the question, admitting that there was a problem, not with the de®nition of noun but with the classi®cation of the word in question: B: A noun is er ± it's a descriptive word for an object I: So it's an object or ± 122

Parts of Speech and Grammar

B: I: B: I: B: I: B: I: B:

Yeah What about say love is that a noun or what? Well it can be a verb can't it? Or if I say em ± he's he's a little love or she's a little love? ± what have you got there? Oh yeah yeah you have yeah oh God you've got me there a bit em ± Or hatred? Yes I know where you're driving now I know where the direction you're forcing me in and I don't like it it's not doing my head any good em ± what is a noun? yeah it's ± Do you think hatred's a noun or not? I suppose it is really isn't it ± but em oh dear yes you've got me it doesn't it isn't necessarily an object is it? doesn't have to be ± no sorry you've got me there I don't know how to answer that question sorry

A at ®rst did not think that love could ever be a noun, a noun for him had to be something tangible, but a little while later, when the word was introduced in the phrase hello love, he considered it a noun. I: Can you give me some examples of nouns ± or tell me what a noun is? A: A noun is a ± a proper noun is a capital city ± say London ± with capital letter ± and it's also an object ± in a briefcase is a noun ± er ± or it can be a person ± can be ± me Bert ± a proper noun again I: Has it got to be something you can see or touch then? A: Yes I would say it had I: What about ± love? What's that? A: I'd say love was a feeling I: Not a noun? A: Not a noun a few minutes later . . . A: . . . hello love is a slang expression for a person I: Is it either a noun verb or adjective? 123

Parts of Speech and Grammar

A: It would be a noun in that sense ± because it refers to a person N and M thought that hatred was a noun even though `it shouldn't be'; for M because it was not `a solid thing', and for them both it was more akin to a verb M: It's used as a noun although it's not a solid thing N: No because if you ± it's used as a noun but it's actually a doing it's a verb isn't it M: It's actually a motion N: You're doing something you're actually hating something aren't you I: So you'd say hatred was a verb? M: Hatred is a noun although it shouldn't be because there isn't a solid thing, but then a lot of nouns are like that, we need a generalisation term N: No you could say hate was a ± hate is a verb M: Hate is a verb but hatred is a noun N: Hatred is a noun yeah M: You can't say I hatred or he hatred N: No I: So it should be a verb you say? Or shouldn't be a noun? M: It shouldn't be ± N: It shouldn't be a noun because you can't ± it's like a chair you can't pick up and look at it and that sort of thing but it is ± well hatred is like a presence isn't it it's there M: Yes it's a motion N: It's a motion yeah but it's de®nitely there it could be a noun but it isn't actually a sort of something you can see M thought that love was an adjective and N thought it either a noun or an adjective: they de®ned a noun as `an item', like a chair or a table: N: Well you could use it [love] as a noun yeah M: Er more adjective I would have said N: Well yes `cause it's a description about how someone feels I suppose really, if you say you wouldn't say hello love to 124

Parts of Speech and Grammar

everyone would you, but it could be a noun because if ± because you were saying to someone that they are a love then they are obviously the love so they could be a noun couldn't they? Could be either. G thought that hatred de®nitely was a noun because it was `a thing': G: What is a noun? ± it's what people call a thing word like a cushion ± something like a glass of beer I: What about what would you say hatred G: Yeah I regard that as a noun yep yep I: So hatred is a thing? G: Absolutely yes I: Why why is it a thing? G: Why is it a thing? ± it just is, it em it's something which you have I mean I mean you must realise that surely ± it's like a hand isn't it or a nose or a pair of teeth I don't know why it is it just is ± it's something which you have em it's not something you can do, well you can say em hand me that watch of course but then you're using well I mean in my parlance you're using ± you've got the right person ± you're using hand there as a verb because I mean essentially I regard hand as a noun and er I regard watch well I don't know how it developed but er oh well probably the earliest one was em the verb actually However, even though he considered hatred to be a noun, he showed some hesitation over love G: the second one as ± hm that's more dif®cult ± that's got to be a noun hasn't it I: Why are you deliberating? G: Er well because I it it for a minute em I was thinking it was an adjective em it isn't it's got to be a noun I: So there hello love is a thing G: Yes [hesitatingly] I: You're not quite sure? G: No I think I am sure yes yes I should say hello darling em or sweetheart mm hello angel ± mm I think it's got to be a noun myself I: Why has it got to be a noun? 125

Parts of Speech and Grammar

G: Because it's referring to a thing a person something absolutely concrete it's like saying hello er Newcastle in a sense you know and er I mean supposing you get to Newcastle station and you say hello Newcastle it's a noun so you say hello love I think it's got to be a noun personally but er I'm open to correction I don't think it's an adjective F like A above, thought that nouns were `physical things'; she also added that a noun was `the main word' in the sentence. There was even less agreement as to what the grammatical differences between the words fancy were (Question No. 6, question 2). A, N and M thought that fancy in what's your fancy could be a verb I: The next two ± what's your fancy ± that's a fancy dress A: Fancy is a ± in the ®rst sentence ± is a slightly slang expression ± it means that ± em `what would you like' I: Mm A: Em ± and the second one ± fancy means ± em ± an unusual type of dress I: Are they nouns or verbs? Either of them? A: I'd say in the ®rst case that fancy could be a verb I: Why? A: Because ± I wouldn't say it was de®nitely was a verb ± and just say that it could be I: Can you say why? Can you explain why? A: Because it's something that ± you could possibly want to do I: Mm A: Fancy dress is a descriptive word ± an adjective N and M were unsure of the word classes for this unit, M thought the ®rst example of fancy could be a noun or verb and N thought it could possibly be a verb: M: N: M: I:

Fancy in the second bit is a in a dress is a descriptive . . . . . . It's an adjective The ®rst one . . . Is it a noun verb adjective? 126

Parts of Speech and Grammar

N: Em I don't know M: I suppose it would be a noun if anything N: No no it wouldn't, it could ± I don't know what that is actually, once you'd got it it would turn into a noun because you'd have it there M: God knows what you'd call it I mean if you ± N: Well if you say what d'you fancy to eat and what d'you fancy to drink M: That's what d'you like basically N: Yeah ± I don't really know what ± M: It's s'pose it's more a verb than anything N: It's a description M: `I fancy therefore I want' N: Yeah so you're saying it's a verb yeah ± I wouldn't say it was de®nitely a verb, I don't know what it is, it's more a verb than anything else F and H thought that fancy in what's your fancy was a noun and in fancy dress an adjective. The majority of my informants were unable to state the difference between lovely and lovingly (question 3); A claimed they were both adjectives: I: A: I: A: I:

That's a lovely picture Desc- it's an adjective He looked at her lovingly It's an adjective again So why there do you say lovingly ± and there you say lovely? ± you say they're both adjectives? A: That's a lovely picture ± lovingly ± er I suppose that's how we use them ± in that text ± there could be other reasons ± but that's the way we've always done it J only commented on the meaning of these sentences without ever mentioning the `word classes' involved J: Well if you put ± if you group them together they're all likeable likeable words you know all words I mean er er 127

Parts of Speech and Grammar

I: Well do you use lovely instead of lovingly or why? J: Well if you describe an object or a thing that you look at and you think it's lovely ± describing a subject object and you think it's very nice it's lovely in fact it's quite ± and then you could look at something again because you possibly may ± this is lovingly ± at you possibly may look at the it at something you'd like to have for yourself so you look at the object subject lovingly and it's something you'd like to have but it's not yours I: So how do you know when to put the `l-y' at the end of love and -ingly at the end of love? J: . . . Em ± describing something that is beautiful ± it's lovely and a yearning or a longing And E thought that only lovingly was different from the other love words of Question No. 6, since it referred to `lust'. And as with J above, E only commented on the contextual meaning E: I: E: I: E: I: E: I: E:

That's just an expression innit really ± I love dancing What about hello love? That's the same innit really? ± just bein' nice That's a lovely picture That's a lovely picture innit ± not like this one `ere [points] I dunno ± I don't really use the word that much lovely He looked at her lovingly Oh that's different innit ± that's that's they same as em lust innit ± really So you think those three words ± the ®rst three loves are similar and the fourth one isn't? Hmm

Although some of my informants (D, G, H and M) identi®ed lovingly as an adverb, others claimed only to have forgotten the name of the term involved: I: He looked at her lovingly B: at her lovingly ± oh mush ± that's another good one isn't it mush that's very much in this area ± oh dear letting myself 128

Parts of Speech and Grammar

I: B: I: B: I: B: I: B:

down badly here ± that's a lovely picture he looked at her lovingly ± em hang on how would I describe that? I forgot the name of the word now how to describe it being used in that way Well can you give me some other words that would be similar to that? ± to lovingly? In the way you mean? Mm Oh yeah well em she dances divinely yeah? Can you explain those sort of words? Well it it is it's descriptive of course but it's not really an adjective it's something else but I forgot the name of the word now Why is it something else ± it doesn't matter if you don't remember the word Well because an adjective is normally describing a noun yeah or accentuating a noun should I say whereas there it's not it's describing a verb or accentuating a verb well yeah? yeah

When asked whether he could de®ne `adverb', G, after some consideration, de®ned it as an `adjectival phrase' G: Em he looked at her lovingly mm that's a very dif®cult question it's a sort of adjectival phrase I think it's a phrase you can't express in an adjectival ways but which you postpone postpone usually not always I think em after the verb ± post pose I suppose they would say nowadays don't they after the verb em no apart from that I don't think I can no D said it was D: . . . a word that quali®es the verb ± it adds a kind of re®nement to the overall meaning of the verb and it it ± er ± has the function of re®ning the er ± idea that is being conveyed And M surprised herself in her explanation of what an adverb was 129

Parts of Speech and Grammar

M: That's em a way of describing how you do something, it's not he is green but then again it's not he is going but he is going green ± God knows where that came from Despite the fact that he was unable to de®ne adverb to his own satisfaction, or to state conclusively whether hello love was a noun, G still felt that knowledge of such categories was `very helpful': G: Em well I think it's very helpful I think this is this is the important thing that em I: Helpful for what? G: Well just to sort out one's de®nition of grammar em I know it's not the er fashion nowadays to talk about nouns and verbs and adverbs and so on and people think that it's very crude simpli®cation of what things actually are, nevertheless I think that it's very helpful . . . I: Does it help you to communicate whether you know something is a noun or a verb? G: Yes I think it does yeah I think you know er when you say I have a watch you say well you know that's a noun but then watch your language you're using it as a verb I: But don't you think that people who didn't know those de®nitions if one person said I have a watch em they would still understand ± whether they knew it was a noun or a verb G: Well in theory I think that's true but in practice I don't think it is true, I think in in practice er yeah in theory it's true in practice it isn't, I think in practice you need to have those de®nitions no matter how crude they are and em I would defend that to the last ditch I think and you can use ditch as a verb if you want to ditch your boyfriend ± er fair enough okay but I think there are those that this sort of thing er can be taken too far and those those awful old-fashioned de®nitions are very very useful . . . the idea the idea of attaching labels to things is not terribly important in a sense but I think we have to categorise things don't we, at least I do I have to have things categorised and it seems to be very helpful in spite of the fact that you know if if er linguists 130

Parts of Speech and Grammar

now feel that it's wrong, I still think it's very helpful to categorise things somehow because you've just got to put things in certain sorts of boxes in order to say you know this is essentially a verb but it may be used as a noun and er that seems to me the most helpful and useful way of going about things that's all and if people don't like that well you know they can ®nd some other method and they haven't yet [laughs] ± c'mon Trudgill ± we're waiting c'mon M, as noted [p. 124 above], in a similar fashion claimed that `generalisation terms' were often needed. D, appeared to be putting forward a Lockean view of communication when he claimed that such terms were in fact a re¯ection of the ideas one has, which in turn re¯ect processes in the world. When asked what nouns and verbs were he said that `a noun is a linguistic vehicle which conveys ideas' and that `verbs also convey linguistically ideas'.: I: What do you mean by `convey linguistically ideas'? D: Em ± they are ± em nouns and verbs in the structure of language ± are the means by which we ± er ± convey our ideas from our minds to er ± another person I: What ± what is the difference between then a noun and a verb? D: Well ± the ± the ideas ± er ± require ± er ± a complex system to convey them properly ± and the structure of that system necessitates different elements ± which nouns and verbs are I: So can you give me some examples of nouns and some examples of verbs to show how you distinguish between a noun and a verb? D: Yeah ± it would probably go back to the idea that is in your mind at a particular time ± it might ± the idea might centre around say two particular concepts which are best understood by the application of nouns say dog and cat ± right ± but ± the ± er ± complete idea in your mind may be that of a dog chasing a cat ± right ± now ± the nouns have supplied the two basic elements in the idea ± that is the `dog' and `cat' ± but ± without a verb the structure of language won't 131

Parts of Speech and Grammar

I: D:

I: D: I: D: I: D: I: D: I: D: I: D:

properly ± em ± won't properly convey the complete idea so you supply the verb which is chases But you still haven't told me how you would distinguish between a noun and a verb ± what the difference is ± how d'you identify a noun and a verb? Well ± grammatically the noun is the ± is the word that is applied to ± em ± to objects or ± Em ± to ± er ± abstract ideas whereas the verb is the ± er ± word required to em ± convey ideas of interactions between objects Okay which clearly is a re¯ection of what happens in the world around you that objects are always interacting with each other If you say ± em I dwell ± use say the word dwell Em Is that an interaction? That's an interaction if you say I dwell because you ± it means you dwell somewhere ± You're in the world ± that's what it suggests to me . . . Okay ± what about weighs in John weighs ten pounds ± what is weigh interacting with? Em Is it interacting? Yes it is because ± em ± it's interacting between the subject of the sentence which would be John and the abstract idea of weight ± in this case a very speci®c weight . . . . . . if I just say birds ¯y John swims Yeah ± in that case the interaction is still taking place ± em ± though in the em ± nature of that interaction isn't as well de®ned as it would be in a sentence say that is capped by another object ± another solid object ± in the case of John swims or John ¯ies we can put John in the context of either the air which is a very general idea or in the context of water.

D's views as expressed here are also reminiscent of the Port Royal grammarians who thought that the `parts of speech' were a necessary consequence of the natural manner in which we express our thoughts. D subsequently told me that he had read neither Locke nor the Port Royal grammarians. In addition D is 132

Parts of Speech and Grammar

here using a familiar pedagogic claim that such categories were essential to `convey the complete idea'. All the de®nitions of noun given by G were of a notional kind and, according to him, were extremely useful for, among other things communication (p. 130 above). Likewise M, who also spoke of `what nouns are used to refer to in the outside world' thought that `generalisation terms' were needed (p. 124 above). Apart from D (p. 131 above), who thought that parts of speech re¯ected mental ideas, the majority of my informants appeared to believe that the assignment of such categories to words was rather chaotic. And as seen here, there was very little consensus on the interpretation of the parts of speech. Whereas some labelled love in I love dancing a verb, A thought it was an adjective because it describes how one feels about dancing (p. 121 above). Love in hello love was considered by a number of informants a noun but N and G hesitated over whether it was an adjective, N because it was a `description of how someone feels' (p. 124 above), and whereas a few thought hatred was a noun, N and M thought it was more like a verb (p. 124) because it was a motion. N, M and A thought that Fancy in what's your fancy was a verb (pp. 126±127) because it was something `that one could possibly want to do', and A thought lovely and lovingly were both adjectives (p. 127). This variation is hardly surprising since grammatical terms require explicit instruction as to their use: we are not taught the word class for each and every word but rather for a `typical' sample of words in certain well (easily) de®ned contexts, a fact recognised by Wittgenstein. If you were unable to say that the word `till' could be both a verb and a conjunction, or to construct sentences in which it was now the one and now the other, you would not be able to manage simple schoolroom exercises. But a schoolboy is not asked to conceive the word in one way or another out of any context, or to report how he has conceived it (Wittgenstein 1958: p. 175e). Even though the majority of people living in twentieth-century Britain can use such terms as nouns and verbs, this does not corroborate the fact that such terms have any independent meaning 133

Parts of Speech and Grammar

divorced from a speci®c context. Even though Sapir pointed out the cultural relativity of the parts of speech, claiming that our conventional classi®cation of words into parts of speech is only a vague, wavering approximation to a consistently worked out inventory of experience. We imagine, to begin with, that all `verbs' are inherently concerned with action as such, that a `noun' is the name of some de®nite object or personality that can be pictured by the mind, that all qualities are necessarily expressed by a de®nite group of words to which we may appropriately apply the term `adjective'. As soon as we test our vocabulary, we discover that the parts of speech are far from corresponding to so simple analysis of reality. We say `it is red' and de®ne `red' as a quality-word or adjective. We should consider it strange to think of an equivalent of `is red' in which the whole predication (adjective and verb of being) is conceived of as a verb in precisely the same way in which we think of `extends' or `lies' or `sleeps' as a verb (Sapir 1921: 123±4). The `part of speech' re¯ects not so much our intuitive analysis of reality as our ability to compose that reality into a variety of formal patterns . . . Each language has its own scheme. Everything depends on the formal demarcations which it recognises (ibid.: 125) he still appears to assume that speakers of a particular language all have a common understanding of grammatical classes. The responses recorded above, as we can see, cast serious doubt on such an assumption. Grammaticality Judgements of Stigmatised Units A general consensus among sociolinguists is that language is not homogeneous, but rather displays a `patterned heterogeneity' ± it varies according to setting and the social identity of the speaker. The major trend of sociolinguistics has been to isolate a linguistic 134

Parts of Speech and Grammar

variable and to correlate such a variable with social/situational factors ± thus even though the pronunciation of singing can be ['sõÎõÎ] or ['sõÎõn], the choice between these variations is predictable ± one pronunciation shows, for example, more `prestige' than the other: We may de®ne a sociolinguistic variable as one which is correlated with some non-linguistic variable of the social context: of the speaker, the addressee, the audience, the setting, etc. Some linguistic features . . . show a regular distribution over socio-economic, ethnic, or age groups, but are used by each individual in more or less the same way in any context . . . More highly developed sociolinguistic variables (which we will call markers) not only show social distribution over socio-economic, ethnic or age groups, but are can be [sic] ordered along a single dimension according to the amount of attention paid to speech, so that we have stylistic as well as social strati®cation (Labov 1970: 283). When a sociolinguistic marker rises to `overt social consciousness', it becomes, as Labov believes, a stereotype and often the strati®cation [hierarchical ordering in terms of, for example, socio-economic or age groups] of a subjective reaction to a (socio)linguistic variable matches that variable in behaviour (ibid.: 293). This near isomorphism leads to the Labovian principle `that social attitudes towards a language are extremely uniform throughout a speech community' (ibid.). But this leads to more dif®culties than beset structuralist linguistic theory, requiring a threefold process of describing `attitudes', identifying a `linguistic variable' and circumscribing the speech community. From the results of my empirical work, it was clear that none of my informants shared social attitudes towards a language, even those who were the same age (A, O; C, R; D, Q; P, T), attended the same school (E, P, T, V, and W), lived throughout their life in the same area (B, E, P, O, T, V and W), belonged to the same profession (A and F), were postgraduate students at the same university (C, D, Q and R) or were engaged at University in a `similar' ®eld (G and Q). 135

Parts of Speech and Grammar

Many informants who referred to the correlation between a `stigmatised' word and social class differed widely as to their judgements. For example, when commenting on was in we was only playing and aint in he aint a pig (Question No. 4), C drew a distinction between a `grammatical error' and the speech of uneducated people; for him, one could easily ®nd educated people saying aint, whereas one would not usually ®nd `the educated' saying we was: C: . . . we was only playing ± that's ungrammatical ± he aint a pig er er ± er well aint is just an ignorant equivalent of isn't . . . I would say that aint is used is a kind of lazy form of isn't whereas was isn't a lazy form of were ± I mean it's it's just a kind of grammatical error committed by people who who don't know any better whereas aint you often get perfectly ignorant people saying aint perfectly educated people saying aint ± don't think you get perfectly educated people saying we was except putting on a voice when they say it But H had different opinions from C, drawing a distinction between a word used `incorrectly' and a word that was `bad', concluding that aint was always a `bad word': H: . . . was like I don't know I would I would say that was improper English it should be we was only playing but was just means that we were playing before not currently or in the future I: Is it the more or less the same as saying he aint a pig? H: . . . was I think it it just should be switched around with were it's not it's not a in inc- like a bad word it's just probably not used correctly in the sentence I: But aint is always a bad word? H: Yes Q thought that both sentences were `ungrammatical' and spoken by the `uneducated'. He also thought that we was would be spoken by the `lower class'. R considered we was `wrong' and 136

Parts of Speech and Grammar

aint `slang'. For H and G, the use of aint was more taboo and for H, especially in its written form since he did not class it as a word, even though it could be found in a dictionary. H: we always learned in school that aint is not a word is improper English and that we shouldn't use it in formal writing and I'm not sure where the word came from but . . . I: You were told it wasn't a word? . . . H: . . . We were told it was not a word but we could still ®nd it in the dictionary and the dictionary said it was slang for is not ... . . . we were always told not to ever use aint it does not belong in writing G however, only spoke of the `upper classes' not using aint: G: Because people from the up- em the upper classes don't use it I: The word aint? G: Yes not nowadays ± they may have done in the past but they don't now M also like C above spoke of `lazy' words but was more explicit in her argument: M: That's just bad grammar because people can't be bothered to speak properly ± and he aint a pig is the same thing . . . . . . It should be he isn't a pig and we WERE . . . . . . It's just a lazy version I mean ± as a whole I don't use it but I must slip out with it now and again ± and with other things it's even quicker basically N: Yeah because it's er I'd normally say isn't I think it's just as easy to say isn't than aint really it takes about the same time to say doesn't it [it is unclear whether this is a disagreement with M's comments above] B produced the following response, referring not only to social class membership but also to the characteristics and geographical location of certain people: 137

Parts of Speech and Grammar

B: Appalling ± we WERE only playing I: What sort of person would say that? B: Engel®eld Green [a village in Berkshire] ± home counties de®nitely ± home counties people of em ± solid working class not wishing to be incredibly unkind or anything like that ± and also people who don't think ± people who know better I'm sure but have allowed themselves to slip into bad habits However, in response to the pair my husband and me/I (Question No. 4, questions 6 & 7), he produced a subjective evaluation: B: Well number seven's right and number six is rubbish I: Rubbish? B: Yeah ± em I'm not going to say my husb- my husb- ± er if we were really to be using that at the beginning of a sentence ± em ± my husband and I were going to do something so we can't say my husband and me were going to something or are going somewhere or whatever, can you? ± not if you're speaking proper English anyway ± not what I consider to be proper English In contrast, G only wished to speak of the correlations of `grammatical errors' with social class membership. He professed that, as a linguist, he did not use such terms as `correct' or `incorrect'. However, it is hard to see the difference between speaking of something as `a grammatical error' or `grammatically incorrect'. This is a point discussed in more depth by Cameron 1995, Chapter 1. G: . . . we WAS only playing oh I see we're getting to grammatical sphere now, well I would take that to be a grammatical error I: What kind of person would say that sentence? G: We was only playing em em some of the lower-middle classes ± all right ± not from the middle-middle classes or the upper-middle classes, someone from the lower-middle classes or the lower classes we was only playing he aint a pig 138

Parts of Speech and Grammar

I: G: I: G: I: G: I: G:

now that's got to be a lower-middle class sentence I think really hasn't it . . . My husband and I ± yes that's just like the Queen yes em well my husband and me would be a lower-middle class sentence er sorry or a lower class a lower class statement Do you think it's incorrect or ± I'm not saying it's incorrect it it's it's something which is used by er lower classes and lower middle classes but is not used by middle middle class people, I wouldn't use it The it's not grammatically correct? I don't well er I don't use those terms ± incorrect incorrect you know but it ± Why not? Why not? Mm Because as a linguist I'm not er ± I can't use er I'm not allowed to use that sort of statement as I never do but it is used by em by lower middle classes and lower classes em my husband and I of course is er a middle-middle class uppermiddle class upper class phrase

But Q, the postgraduate linguist, thought that my husband and me was `wrong' `uneducated people say it'. And A invoked pedagogic instruction and, like G above, the in¯uence of the Queen's speech to justify I: A: . . . my husband and me it's the archold expression that we all know we're not supposed to say, that we were taught not to say at school, but everybody does say it and the next one my husband and I is supposed to be the correct way of saying that expression ± because as as I say ± that's how we were taught at school and the Queen says it and that's the in¯uence on our society But according to V and W, my husband and I is a phrase which exclusively belongs to the Queen: V: You say my husband and me not my husband and I, my husband and I is Queen's English 139

Parts of Speech and Grammar

W: The Queen's the only person to say it O said that the phrase my husband and I reminded him of the Queen, but that her speech was not necessarily `correct pronunciation': O: . . . the second one I always think of the Queen there . . . I: . . . what would you use? would you say my wife and me or my wife and I? O: The wife and I actually, if I said it like that er don't use me that much I: Are they both okay to say? O: Er I suppose in correct pronunciation not it isn't you've gotta say my husband and I you really say my husband and me as far as I can remember. E thought that me was `absolutely grammatically correct' but that since he was `brought up to say I', that is what he would say. Unlike G, three of my informants spoke of contexts in which me would be acceptable. C spoke of a context in which it would be `absolutely correct' to say me: C: . . . well my husband and me is er ± well I think it's ugly and it's normally wrong em ± unless of course it's used in the accusative if er ± if you said Hayley came to visit my husband and me then it would be absolutely correct because me is the object of the visit em ± nine times out of ten if people say my husband and me they mean my husband and I Here, C did not feel it worth explaining how someone can mean `my husband and I' while saying my husband and me and vice versa. This informant then continued to speak of the unpleasantness of the sound: I: You said it was ugly ± what d'you mean by ugly? C: Doesn't sound very nice I: It's just the pronunciation? 140

Parts of Speech and Grammar

C: Yeah well `tis well yeah yeah my husband and me ± I suppose because you don't often see it ± it jars on the ears but it's not necessarily wrong M at ®rst exclaimed my husband and I was `bad grammar' but soon after both N and M spoke of contexts in which me could be used; unlike C however, they were unable to account for the correctness of the phrase: M: THAT I would class as bad grammar N: Yeah I would I would say my husband and I . . . M: . . . it depends on what's before and after them, I mean if you had em we went ± there was [sic cf. p. 137 above] four of us went my mum my dad my husband and me it could be alright N: That'd be right yeah M: But you'd say my husband and I went with my mum and dad N: You can put me in it sometimes ± there's different ways of using sometimes you have to use I sometimes you have to use me but I'd say with those two that the second one was right M: Yeah as it stands the second one would be right E also produced a sentence in which me would be the preferred pronoun; although in this case the reason for his choice was somewhat unclear: I: The next two you can take together my husband and me ± my husband and I ± what would you say about me and I there? E: Dunno I: Which one would you say? E: My husband and I I: You'd say my husband and I? E: Yeah ± or my wife and I and wouldn't say my wife and me I: Why not? E: Unless she was going on holiday I: Why unless she was going on holiday? E: Or going somewhere I: Why? E: Dunno ± it just sounds better 141

Parts of Speech and Grammar

F preferred I in a similar context, giving a somewhat unclear reason for her choice of pronoun F: . . . my husband and I sounds a lot nicer I: Why? F: my husband and me points to me rather than I ± it's saying me rather than as in together P also thought that I sounded better and that the only difference was who was mentioned ®rst: P: It's just putting yourself ®rst ± mm it's just saying somebody else's name ®rst, it doesn't matter whether it's me or I does it, I sounds better though I: Why? P: Dunno why H thought that my husband and me was `grammatically incorrect' but it `still gets the idea across' and J and D did not think that either of my husband and me/I were incorrect J: Well over the years they were both correct em people have used both and I I see nothing wrong in either of them I: So you'd say they're both grammatically okay? J: Yeah D: Well actually I wouldn't ± wouldn't separate the two ± I think that they actually mean the same thing but that em ± the latter sentence regarded by ± er certain people as being ± er ± grammatically and more conventionally correct ± though I don't agree ± I think that they mean the same thing D did think that he was mad at me was incorrect. The reason he gave was that the verb plays a more important part than the pronoun: D: Em ± that's er ± was there is grammatically there the wrong part of the verb to be using though it's quite a common mistake . . . the mistake there is in the verb which plays a 142

Parts of Speech and Grammar

more important part in that sentence than the pronoun in sentences six and seven [my husband and me/I] ± I don't believe they harm the sense J also thought he was mad at me to be incorrect ± his opinion on aint was that, unlike D who thought it a `regional variation', it was also incorrect being ungrammatical and not `proper English'. His justi®cation for this comment was similar to A's (p. 139 above) in referring to teaching at school: J: Well it's not correct English well it's not only aint isn't ± well it's not correct English I: Why isn't it correct? J: Well grammatically it's not correct I: Why? J: Because my English teacher told me so a million years ago This was also his reasoning for saying that mad was incorrectly used in he was mad at me (see Ch 4, p. 113 above): I: Did your English teacher tell you you can't say he was mad? J: Well it's grammatically he she er would say it's not correct to say that though you can understand though people say it or people say many things that you understand but you know it's not correct The majority of H's judgements were based on pedagogy, either at an elementary level (see p. 137 above), or at senior level involving essay writing and marking : I: . . . is there a difference between saying my husband and me, we was only playing, he aint a Pig? H: It's just whoever makes up the rules for proper grammatically correct English . . . . . . I mean everyone every you say any one of those sentences and everyone will know what they mean . . . But if someone was grading their writing then they would say that that was incorrect 143

Parts of Speech and Grammar

A, referring to aint (Question No. 4) claimed that apart from Americans, the word was used by `people with a limited vocabulary' and D spoke of a regional variation: D: Er ± this is a kind of ± I would have thought that this is a kind of slang verb in a way ± well maybe not slang but maybe it's em ± maybe it's just a kind of regional variation for isn't ± em ± yeah I think that's probably a better form of explanation On these questions from Question No. 4, F did not mention the grammaticality of the sentences at all; only the way such sentences `sounded' or who would say them: F: . . . we was only playing ± `we were only playing' I: So what about the word was there? F: It doesn't sound right em ± well past tense perhaps unless it was slang then it wouldn't ± he aint a pig that sounds whinny and horrible I: Why? F: Em ± `cause it's a slang word and it just doesn't sound `cause if you'd say it in a normal sentence he's not a pig or he aint a pig is very ± I: So who would say it ± what sort of person? F: A child I: A child? F: I would think yeah em ± or perhaps somebody who's substandard in education . . . Likewise, E usually only gave his interpretation of the meaning of such sentences ± the only `linguistic' analysis he gave was of the word aint which he considered an abbreviation: E: we was only playing ± what would you say about that? ± it's a sympathetic word innit I: What d'you mean by that ± why? E: It's like when you've done something wrong innit ± when you was a kid ± we was only playing 144

Parts of Speech and Grammar

I: Okay ± he aint a pig . . . E: . . . Aint is just an abbreviation for innit ± [pause] isn't ± is not I: Abbreviation? E: Must be ± he aint ± he isn't ± just the same O expressed only what such sentences meant, and devised some situations to `explain' their uses: O: We was only playing ± was that's the ± to me that means couple of kids playing and one's broken a window and says we was only playing broke a window that's all ± he aint a pig that sounds like a girlfriend sticking up for John at the top, but there again John is a pig, she says he aint a pig they both deserve each other anyway The ®rst half of this chapter has, in part, questioned the views of those psychologists who claim that children, at a certain age of development, acquire a complete, or near-complete, understanding of such grammatical terms as noun, verb, etc. (see Berthould-Papandropoulou 1978, Tunmer, Pratt and Herrimann 1984)), since my informants showed very little agreement as to the meanings of such terms, even those informants who were highly literate/educated. The second half of this chapter has agreed with J.R. Ross's misgivings about the belief that speakers of a language generally agree among themselves as to the degree of grammaticality of test sentences (Ross 1979: 128). We can therefore see the dif®culties inherent in speaking of the co-variation of social class membership and grammatical speech behaviour, since one factor in the de®nition or identi®cation of members of certain classes is their use of `ungrammatical' language. But my informants have clearly demonstrated that what form is `stigmatised' is neither an uncontentious matter, nor something that is stigmatised by all for the same reasons: for instance G thought aint a form not used by the upper classes (p. 137 above), whereas for C it was a form used by `perfectly educated people' (p. 136 above) and for A it was a form used by people `with a limited vocabulary' (p. 144 above) whereas for J it 145

Parts of Speech and Grammar

was `incorrect' and `ungrammatical' (p. 143 above). However, it is not clear whether educated people use incorrect and ungrammatical forms or whether the upper class and `people with a limited vocabulary' are mutually exclusive. We are therefore in a good position to cast doubt on Labov's axiom that the correlate of regular strati®cation of a sociolinguistic variable in behaviour is uniform agreement in subjective reactions towards that variable (Labov 1970: 293). Although the majority of informants above had some knowledge of grammatical terminology, such knowledge was by no means uniform among all the informants. Furthermore, some informants appeared to have no understanding either of the grammatical terms or concepts themselves or even in one case of the word grammatical itself (see O p. 122 above). However, it did appear that the notions of `right' and `wrong', `correct' and `incorrect' featured prominently in the layperson's characterisation of words (see O, A pp. 139±140; C p. 140 ; D p. 142; J, H, p. 143).

146

6 Folk Characteristics of Words

This chapter is concerned with laypeople assigning various, often subjective, characteristics to words. Such characteristics include the use of such terms as `bad', `nice', `descriptive', `wrong'. This way of talking about words seems to be, from a lay point of view, quite important, often being used as the basis for identifying a speaker as, for example, arrogant, stupid, rude, silly or clever. Some of the properties discussed below are `folklinguistic' for the reason that professional linguists steer clear of such terms and labels as they only get in the way of a `scienti®c' description of the language. The scienti®c study of language has convinced scholars that all languages, and correspondingly all dialects, are equally `good' as linguistic systems. All varieties of a language are structured, complex, rule-governed systems which are wholly adequate for the needs of their speakers. It follows that value judgements concerning the correctness and purity of linguistic varieties are social rather than linguistic (Trudgill 1995: 8). Consequently, one of the ®rst axiom's of orthodox linguistics is that `linguistics is descriptive not prescriptive'. Folk characterisations, although not requiring the same degree of explicit formal instruction as the `parts of speech' classi®cations or spelling conventions, nevertheless have other cultural determinants. As well as noting these folk attributes which are assigned to various word-forms, I will also be investigating 147

Folk Characteristics of Words

various cultural in¯uences which may have inspired the choice of a particular label. Question No. 3 was an investigation into the categorisation of particular words. My informants were asked to both `comment on' the pairs of words and, where appropriate, to give a reason for their choice of description or label. Yet again (see p. 38 above) the fact that a response may have been produced which may not have been consciously formulated, let alone articulated in the absence of such an interviewer/ee situation, is a general phenomenon of communicational interaction and does not therefore detract from, nor necessarily misrepresent, the importance of the recognised cultural in¯uences on subjects' explanations. Bad Language `Bad language' is a topic which frequently attracts attention from the media (Davis 1998). Apart from reports of such language being used by famous sportsmen (e.g. The Independent, 10 December 19871), journalistic articles on this subject are usually of a speculative or critical nature: `What is it about this word, four tiny letters that, in combination, make grown men ¯inch and women cringe?' (The Guardian, 2 November 1989); there does seem to be an increased ± and increasingly ± number of cases of what is generally referred to as `bad' language, and it's starting to creep into the schedules much earlier in the evening (The Evening Standard, 21 November 1989). Thus, it is often the language used in public which causes more offence than the troublesome situation being described. For instance in October 1998, Richard Branson was asked to 1 In 1987, there was a confrontation between Mike Gatting, the captain of England's cricket team and Shakoor Rana, the Pakistani umpire. The incident culminated in Rana allegedly calling Gatting a `fucking cheating cunt' (Davis 1998: 283).

148

Folk Characteristics of Words

apologise for his repeated use of the verb `to fuck up' on a radio broadcast in connection with his railway company's inability to deliver delegates to a Labour Party Conference on time. It is as if we are being told that we should be more offended by Branson's linguistic performance than by his trains' poor service performance. It is thus hardly surprising that such `bad' language is familiar to the general population. But the `linguistic facts' are far more complex than the accounts given would have us suppose. Question no. 7, asking informants to comment on the words blimey and shit elicited the most detailed responses, both by those who had below the current legal minimum of formal education and by those who were university educated. Also, differing from comments often made about other pairs of words, my informants did not generally ask what they were supposed to say or what was meant by the question `what sort of words are they?' Some even produced a judgmental response prior to labelling such words: B: . . . great ± lovely stuff G: . . . blimey and shit are interesting E: Now you're talking my sort of language The properties of these words were not characterised with uniformity, nor was there any consensus on the de®nition of `swearword' or `slang', the two most popular labels. Some other terms mentioned were `®lthy language' (B), `silly word from a not very intelligent Londoner' [blimey] (Q), `both dirty words' (S), `cuss words' (H), `both blaspheming words' (K), `expletives' (G) and `both expressions' (W). F, T and J thought they were both `slang'; for B blimey was a Cockney/slang expression and shit `slang' whereas for C, D and P they were both swearwords; the difference according to D was that `one is obscene and the other isn't'. A said that shit was a swearword and blimey an exclamation `but they're both slang'. O thought that only shit was a swearword. Some were unclear as to the application of a certain label to such words, although the words themselves were familiar: 149

Folk Characteristics of Words

N: Well it's [shit] a swearword but it's not ± some people would class it as a swearword and ± it's used so often that really it doesn't seem like that E: I: E: I: E: I: E:

Shit's shit aint it What sort of word is it? It's not really a swearword is it? What is a swearword? Eh? Can you give me a word that is a swearword? No not really

Moreover, according to two of the informants it was neither a necessary nor a suf®cient condition for a `four-letter' word to contain four letters nor for a `dirty word' to be unacceptable. I: What is a swearword? O: A four-letter word even though bloody isn't a four-letter word S: . . . both dirty words, the ®rst is acceptable though Although, and because, such terms are essentially of interest to the general (TV-watching, newspaper-reading) public, they are not treated in mainstream linguistic inquiry, but are rather relegated to such sub-disciplines as sociolinguistics (see Trudgill 1983; Andersson and Trudgill 1990) and sub-areas as popular discussions of language (Montagu 1968, Partridge 1972, 1989, Howard 1984, Michaels and Ricks 1980, Ricks and Michaels 1990). The covert prohibition on the use of such terms is often explained in such writings by appeal either to the reference of the words: `Four letter words are deemed obscene because they refer to an aspect of life that has long been considered ob caenum, that is ®lthy' (Montagu 1968: 301) and/or to social values: `the type of word that is tabooed in a particular language will be a good re¯ection of at least part of the system of values and beliefs of the society in question' (Trudgill 1983: 29). Trudgill acknowledges that explaining such taboos in terms of the ideas or social values expressed by the particular words does not account for the full range of facts as witnessed in the perfect acceptability of synonymous words. He concludes that the 150

Folk Characteristics of Words

reaction to swearwords is due `to an irrational reaction to a particular word' (Trudgill 1983). However, it was more common to ®nd my informants believing that such words were taboo because they re¯ected taboo ideas or things: D: Er ± well ± em just swearwords really I: Both of them? D: Yeah ± I think so ± yeah ± though one is obscene and the other isn't I: What do you mean by swearword and obscene? D: Em ± a swearword is a kind of a ± an oath or a curse on someone or something ± er and by ± er and by obscene I would mean the application of a word er ± the application of a word that is taboo or that represents an idea in the context of swearing I: What makes a word taboo? D: Really the same thing that makes an idea taboo ± em ± because I mean words are only representations of taboo ideas anyway ± that's that's a kind of anthropological question or a psychological one C: . . . the word fuck describes the act of sex and you'd tend to ®nd that em ± you know the stronger ones the ones you know to go up from fuck you get a kind of a a specialisation of it like motherfucking and things like that er I mean that obviously describes the a slightly more shocking thing em ± I don't know what blimey refers to A frequent academic explanation of the uses and users of such words is that because they are `sub-standard' language, they are employed by members of the lower classes, in particular male lower classes. Trudgill (1983), for instance, says that since swearwords are `nonstandard' language, women being more conscious of propriety and upward mobility try harder to avoid using such terms, but that since social class is just as an important sociolinguistic variable, working class men and women use more stigmatised forms. For Risch, in her survey of derogatory terms that females use to refer to men, the standard/non-standard 151

Folk Characteristics of Words

distinction is `more appropriately applied to the contrast between public versus private discourse than to that of the speech patterns of women versus the language use of men' (Risch 1987: 353). `It could be that nonstandard linguistic varieties carry some value of covert prestige for communities of young females' (ibid.: 357). Only one of my informants mentioned any sex bias in the use of such terms; he did not mention any class correlation but rather a nationality correlate: G: . . . shit as far as I understand it is being used more and more by American young girls as an expression of disgust Some lay interpretations concurred with Andersson and Trudgill's de®nition of swearing as `a type of language use in which the expression . . . can be used to express strong emotions and attitudes' (Andersson and Trudgill 1990: 53). I: What would you say a swearword is? ± can you de®ne one? C: How would I de®ne a swearword? ± er I was going to say it was ± a word you say when you're angry but that's obviously not the case thinking of some people ± that's a dif®cult one ± I would I would just I would say they're words that are taboo in polite society that are usually used in anger em ± yeah that will do For N the difference between blimey and shit was that blimey was not a `shockeable' word: I: N: I: N: M: N:

And what is a swearword? A lot of people use them to shock don't they I suppose What's the difference between blimey and shit? Well blimey's not sort of a shockeable word is it? Blimey's an abbreviation for when they used to say God blind me Yeah that's right I remember reading about that at school ± but it's ended up being just blimey and you say that well I dunno if you did but at school when you were young it's one of the ®rst things you start sort of saying ± as you get older you tend you use to start saying shit 152

Folk Characteristics of Words

Like N above, many informants claimed one bases one's understanding and use of swearwords on what one has been taught or read about them: I: What makes a word a swearword?. . . A: I suppose it's mainly to do with one's upbringing really ± whether you consider it to be a swearword or not I: It's what you're taught? A: Yeah I think so ± the environment that you live in O: . . . they're not words that you get taught at school if you know what I mean I wouldn't ± er if a kid of mine said shit or bloody I think I'd have a go at it . . . Many informants appealed to lexicographical practice to justify their opinions on such words, either referring to lexicographical labels such as `slang' and `swearword' or to the dictionaries' custom of omitting such words (see also p. 63 above). Ironically however, lexicographical and lay opinions generally did not concur. The OED's de®nition of a swearword is a word `used in profane swearing, a profane oath' and `to swear' is To utter a form of oath lightly or irreverantly, as a mere intensive, or an expression of anger vexation, or other strong feeling: to use the Divine or other sacred name, or some phrase implying it, profanely in af®rmation or imprecration; to utter a profane oath, or use profane language habitually; more widely, to use bad language Slang is de®ned as 1a) The special vocabulary used by any set of persons of a lower disreputable character; language of a low and vulgar type b) the special vocabulary or phraseology of a particular calling or profession; the cant or jargon of a certain class or period c) Language of a highly colloquial type, considered as below the level of standard educated speech, and consisting either of new words or of current words employed in some special sense 153

Folk Characteristics of Words

The majority of the dictionaries I consulted indicated the `register' of a particular word, i.e. whether it was `slang' `a use of the most informal kind, unsuited to written English and often restricted to a particular social group' or was `offensive' denoting `uses that cause offence, whether intentionally or not' (COD xxxii). In this particular dictionary, blimey is registered as `coarse sl'. [slang] an expression of surprise, contempt, etc.' and shit and bloody also as `coarse sl.' the former `usually considered a taboo word'. But in the OED blimey is said to be a vulgar `corruption of the imprecation blind me' and shit to be `now not in decent use' (as sb. and v.) but (int.) merely a `coarse exclamation of annoyance or disgust'. For some informants, the difference between slang and swearing is that slang `takes the place of another word' and usually involves the use of `bad language' whereas a `swearword' is not necessarily an indecent or profane utterance: B: . . . I think in many ways a lot of people get it rather wrong ± em ± people think ®lthy language em ± is is always swearing ± to my mind it's not always swearing because a lot of the words which are used in s'posedly ®lthy language are actually slang words I think ± but then again there's a word like bloody ± now that is probably a swearword right? I: What do you mean by a swearword? What . . .? B: Because it's not a slang word ± it's not actually taking the place of another word I: So slang takes the place of another word? B: Yes I: And what is a swearword? B: A swearword is just to is just a word put in for the sake of it ± whereas yeah ± you call somebody a you know you tell somebody to fuck off ± okay well that's that's not necessarily em ± that really isn't a swearword ± it's a slang word I: Can you give me a swearword? B: Well bloody is a swearword as far as I can see ± because it's really it's just a word which is being used for the hell of it rather than taking the place of another word 154

Folk Characteristics of Words

But ironically, this informant appealed to the dictionary for metalinguistic judgements on slang and swearwords: B: Shit? ± well it's just slang for `crap' isn't it ± or `faeces' ± whatever you want to call it I: What is a slang word ± d'you think? B: What is a slang word? ± em ± it's a word that has grown up through the years I suppose em ± which is another word meaning the same as the proper word like obviously shit crap etc . . . . . . What makes a word slang? ± er ± now well ± good question ± em ± er ± I suppose em ± er ± I suppose probably the fact that it's em ± I dunno ± I don't write the dictionaries ± the dictionaries tell us it's slang ± also you know you read a lot of the larger dictionaries and they have a word like shit and then it says `slang for faeces' yeah I: What about swearwords? B: Swearword and slang in some ways the same but not always A referred to the dictionary but came to a differing judgement from B's: A: . . . I mean shit ± it could ± it's not necessarily a slang word ± `cause shit has a meaning I: So slang words don't have a meaning? A: Slang words have meanings but em slang words don't normally appear in the dictionary although some of them do I know ± but I would term it not to be a word or think it not a word if it wasn't in the dictionary O also thought that shit, being a swearword, would not appear in the dictionary: O: . . . swearword's exactly what it is shit and bloody they're they're swearwords you won't ®nd in any dictionary, although shit can be found in the Bible 155

Folk Characteristics of Words

But although such words do appear in the dictionary and moreover are de®ned as being `slang' or `taboo' by virtue of being in the dictionary, which particular words are considered by the layperson as slang, obscene, swearwords etc. are not necessarily in conformity with the lexicographer's judgement. Both my informants here and lexicographers are being asked to consider such words in the context of being mentioned, not used, in order to ascertain the metalinguistic properties of their use. Yet the interpretation and understanding of such a context will be as variable as the context of use itself. Other in¯uences related to literacy were noted in discussions of `bad language'. For example J, who considered the test pair both slang, de®ned slang as `not words you'd use in grammatical English', and G, not passing any evaluative judgements, thought them both `expletives' requiring an exclamation mark after them: I: G: I: G: I: G:

What type of words are these two? Er er well they're what do you call it er expletives Both of them? Yes In the same way? Em I'd have thought so they need an exclamation mark after them blimey em they're expletives I: What is an expletive? G: Em something which one feels er surprised or cross or angry or something like that ± er oh shit or blimey yeah I would put an exclamation mark with those I would call them expletives But for B (p. 161 below) expletives were not words but reactions. P like B (p. 154 above) thought that a swearword was not a `necessary' part of a sentence and that slang was a shorter word which took the place of another: I: What makes them swearwords, what is a swearword? P: Er shit could be used a proper word really, used in a bad term as well I: How does slang differ from a swearword? 156

Folk Characteristics of Words

P: Slang differ from a swearword? Slang's usually something that's shorter or something like that isn't it and a swearword's usually something that's used in a sentence but not really necessary though I: So slang's a shorter ± P: It's usually em kind of a used a word that's used for a different sort of like a different meaning to it isn't it, it's got like two meanings usually or it can be a word that's shortened For D also, slang took the place of another word, but the explanation given was detailed, involving historical and subgroup considerations. Moreover, slang for him did not involve obscenity: I: What is slang? ± what's the difference between slang and a swearword? . . . D: . . . Well slang doesn't have to be ± slang won't be obscene I: So what is slang? D: What is slang? ± er ± I would de®ne slang as ± er ± the replacing of er ± universal or unconventional nouns with ± er ± localised variants ± originally probably slang ± em slang words come from communities that are much smaller than those in which the words are eventually employed ± for instance we possess in our language many words ± er ± American slang ± em ± that didn't actually begin here at all ± that's how I'd de®ne slang J held a similar view: J: blimey is a a is a Cockney word a slang word an abbreviation of corblimey . . . blimey appertains to just one section of of of a country just one section of the public ± cockney ± knowing that blimey is an abbreviation of corblimey O considered the term slang to be a new word for swearing I: What's the difference between slang and swearing then? 157

Folk Characteristics of Words

O: Nothing really nothing, I think slang is just a short term for swearing I mean years ago it was always swearing, slang's relatively new word as far as I can remember, but I'm thirty ®ve I can never remember anyone saying slang until what twenty years ago, it was always swearing before F thought the examples were both slang and that slang was `nice' compared with swearing F: . . . well obviously they're both slang I: What is slang? F: What is slang? not a nice way to talk em I don't know how to answer that I: Are they both slang? F: Yes but one strikes me as being old I: You can't say what slang is? ± can you give me some more ± examples F: what slang? I: Yes F: Yes yes I could I: Go on then F: Do I have to say it Hayley? ± this really isn't nice of you ± blooming heck ± ¯ipping heck ± shit ± fuck ± bugger I: There's no difference between all those? F: I don't think there's any difference between slang and swearing I: No difference at all? F: No because I think they're both detrimental really I: You can't say what it is? slang or swearing? F: Well er no there is obviously a difference slang's more like being lazy with words but it's it's said detrimentally and I think something can be said detrimentally there I: Is swearing being lazy with words? F: Actually no ± I suppose slang's quite nice really ± compared to swearing it is C and R thought that both words were swearwords, the difference being that one was stronger than the other: 158

Folk Characteristics of Words

C: I: C: I: C: I:

. . . well they just strike you as swearwords Both? Yeah Equally? No ± no I'd say shit was a bit stronger than blimey . . . You said em ± shit is stronger than blimey d'you think that there's em a grade from a stronger one to em ± C: Yeah `course there is I: What would the stronger ones be? C: Fuck

O also thought that blimey was not as bad as shit, the reason he gave was that the former was not a swearword whereas the latter was. For ®ve informants shit was just a common everyday word: B: . . . shit well it's just everyday word isn't it ± everybody says it all over the place ± whereas not many people say blimey ± very much a southern thing M: . . . shit's universal ± everybody uses shit for everything N: Yeah shit's universal ± everyone says shit J: Shit is a more common name for excrement I: Do swearwords have a meaning then? ± you said slang words don't A: Swearwords can have a meaning yeah ± I mean like fuck is has got a meaning but em if you say shit you're not em saying it literally em ± you know you're not going to shit yourself ± but ± er ± it's used a lot G agreed with B above in claiming that blimey was not very common but he differed as to the reason why: G: blimey is very old fashioned I think expression which was used em in the forties and has virtually disappeared as far as I know from present-day English usage C's opinion was that any inhibitions on the use of such words would depend on one's audience: 159

Folk Characteristics of Words

C: There are grades of swearwords you only have to judge by the ones people would use in front of people they didn't really know ± I mean we'd all use things like blimey not so many of us would use shit H, the American informant, attempted to distinguish between slang and cussing: H: I: H: I: H: I: H: I: H: I: H: I: H:

Okay they're both cuss words What is a cuss word? . . . What makes something a cuss word? Well I don't know for sure they're just slang words for ± . . . . . . What's the difference between a slang and cuss word? Well I think a lot of the cuss words are slang like shit I think shit can be it it's isn't it like slang for faeces or something? I don't understand what you're saying slang is as opposed to a cuss word . . . . . . Well slang ± slang is a type of word I guess I really don't know I don't know for sure em ± Is fuck slang? I would say it is yes you know when you shout it out in anger you're cussing Ah so if if you're not being angry you're using ± If you like you're using ± hm ± I think it's both a cuss word and slang like if I say I need to go take a shit Mhmm I think shit is slang but I don't think you're necessarily cussing then

The majority of informants who brought up the conditions of use for slang considered such words to be `lazy' (e.g. F. p. 158 above) or not used in grammatical English (e.g. J. p. 156 above) or `incorrect' and `not a proper way to speak' (p. 63f.). A number of other words in Question No. 3 were also labelled as slang or swearwords, or it was thought that, at least in a certain context, the particular word in question could be used as slang. This highlights the fact that not all swearwords are `bad' when being mentioned and/or used: some may acquire such 160

Folk Characteristics of Words

characteristics only in the context of use (see Davis 1998 for an elucidation of this point). D, for instance thought that gazooks (qu. 11 in Question No. 3) was a (non-obscene) swearword and he suspected `that it means God something': I: Do you think it's taboo? D: No it's sacrilegious which is ± yeah ± it's sacrilegious ± it's like a ± it's like a religious form of taboo in a way I: So you would never use that word? D: Oh well ± yeah I mean ± no I wouldn't use that word but it ± it's an old word ± but ± er ± I would use words that are ± em ± sacrilegious J, P and Y thought that the pair ouch and hey (qu. 2) were `(almost) slang words': J: Well well er slang almost a slang type of word P: They're slang words aren't they ± just ± well they're not slang words they're just words you say when you're ± I dunno not real words are they but ± I: Why aren't they real words? P: I suppose they could be real words ± they're just hey I dunno ± hey doesn't really it's just a term of calling somebody isn't it and ouch is just something you do if somebody hurts you or something B thought they were expletives and not real words (cf G p. 156 above) B: . . . they're both expletives I: What do you mean by that? B: Er ± well if somebody ± you know ± let's face it they're not really words they're they're em em reactions D thought that thee (qu. 6) was sometimes `used in a slangy way now' and that stingy (qu. 9) was slang. P also thought that stingy was slang and, although not `the proper word' was `the word that we use': 161

Folk Characteristics of Words

P: Stingy's like a slang word for somebody that's tight isn't it I: And parsimonious means more or less the same as stingy, so what would you say about parsimonious? P: Parsimonious is probably the proper word and stingy's the word that we use ± it's like a slang word for it P also thought that cuckoo (qu. 8) could be used either as a `proper word' or as a `slang word': P: Cuckoo's a proper word but can be used as a sort of like a slang word type you know saying somebody's cuckoo Although O also considered stingy to be slang, he thought that it had been replaced by a different slang expression: O: Oh stingy's again a swear- slang word I would have thought I: If I tell you that parsimonious means more or less the same as stingy what would how would you contrast those two words? Why would you use one and not the other one? O: Well I don't think anyone would use that one I: Nobody at all? O: Well I suppose if you were in court you get some QC'd come out with it and then you'd have to ask them what it meant but if he said stingy which means `tight' as far as I know I'd understand what you meant I mean nowadays you gonna use a slang word you wouldn't use stingy `you tight bastard' in a sentence ± that's it Thus it can be seen that the class of words or expressions subsumed under the term `bad language' is not well-de®ned. Although the informants were familiar both with the `stigmatised' words and with the metalinguistic terms swearing, slang etc., there was little conformity on judgements on their de®nition, application or use.

162

Folk Characteristics of Words

Old Words For the majority of informants, slang and swearwords were generally common, familiar and everyday words. They thus contrasted with words considered by my subjects to be old, specialised and uncommon. Yet again the words thus characterised often differed quite markedly between informants. For example alas and lest (qu. 4) although familiar to all the informants, were considered by some as `old' or `archaic'. F, H, and G thought alas to be more common than lest F: Old English although alas is used I suppose sometimes I: And what sort of word is alas? F: Sorry do you mean the meaning of it? I: No what sort of word is it? F: It's old I: Old? F: Yes ancient G: Alas is an archaic term er lest is also an archaic term in my opinion whereas I would use in case but of course it's still used: alas is an archaic political term so is lest I think em whereas I say I'd use in case em I think lest is passing out of usage em but is still almost in usage I: Are they common words? H: No alas is probably more common than lest ± but I don't know if everyone would know what it means C: Well they don't strike me at all, again a little bit ± lest you don't see very often Others emphasised the fact that such words did not occur in everyday conversation, for H they were more likely to be found written in books, and for M, N and B in literary works: H: Alas well that's like reading books and stuff, that's usually what someone says when they come up with a good idea or something like that 163

Folk Characteristics of Words

I: Alas? H: Yeah I think that's the way it's em lest ± I'm not sure what that means M: N: I: M: N: M: N:

They're very theatrical Yeah Why? Sort of Most people don't use used in everyday conversation In Shakespeare plays Yeah and alas poor Yorick

B: Lest he does something ± alas ± alas ± is really a ± again it's a reaction almost ± isn't it ± alas alack o woe is me ± that kind of thing ± yeah ± lest he does something ± I dunno ± possibly being in some old possibly Dickensian novel ± I dunno how to react to that one actually ± no haven't got much comment to make on that E also used a literary citation while professing not to know the meaning of such words: E: Alas I know him well ± I wouldn't know what they meant really I: Do you know what sort of words they are? E: No not really ± just the English language innit T commented that such words were `Shakespeare' and V said that `they could have the same meaning ± death' and W agreed, citing two quotations to back up his point lest all people die and alas poor Yorick. Q just commented that they were `very literary words . . . archaic . . . not very sincere words'. O claimed that he had only heard lest used once in a war context, and he offered an example of the type of person who would use alas: O: . . . well lest I never use that word lest we forget I: Why? Who would use it then? 164

Folk Characteristics of Words

O: I dunno the only time I've heard it mentioned I suppose is during the war ± when they say lest we forget that's the only thing I can think of there ± alas I can't say I use that word a lot I: Who would use that? O: Some upright citizen I presume, when you gotta say alas you can say O bugger it J commented on the meaning and P inquired whether the purport of the inquiry was to state the meaning of such words and concluded that they were `normal words' J: . . . well well I mean one's a woeful word ± I can't ± well I can't seem to er further lest ± I can't I can't seem to put a meaning to the ± er em for instance I can't see why they're paired I: All right and why is that one a woeful word? J: WHY is it a woeful word? I: Mm J: Well alas is a is a word that means sadness or woeful P: I: P: I: P:

Do you want me to tell you what alas and lest means? Just what sort of words are they? Just normal words Normal? Yeah

Legal and religious Words Hereinafter and witnesseth (qu.3) and amen and thee (qu.6.) were also considered by the majority of my informants to be `old words'. The `register', in the case of the pair of words in qu.6 in the opinion of eight of my subjects (Q,R,T,V,W,X), was said to be restricted to religious contexts (`religion strikes me on both counts (B)), and religious and/or legal in question 3: G: . . . em hereinafter and witnesseth em em biblical expressions I would have thought I: What makes you say that? 165

Folk Characteristics of Words

G: Well -eth means sixteenth century language em and so hereinafter is sixteenth sorry seventeenth century word really er or it could of course be a legal term . . . . . . hereinafter could be a legal term using archaic language but witnesseth has got to be a biblical term because it's got the -eth on it and em unless of course that can be used as a legal term I don't know but they're either biblical or legal terms both of them I'd've thought B did not think that these words were common and appeared to concur, as did M and N, with G's views on the register: B: . . . hereinafter is a lot more common than witnesseth ± but let's face it hereinafter isn't exactly common I: Why not? B: Em ± it's not a word that I hear I: Can you imagine where you would hear it? B: Hereinafter ± witnesseth ± legal? I: Both of them you think? B: No ± hereinafter is legal ± witnesseth religious I: Why do you say that ± what about them that makes you think that? B: What makes me think that? ± witnesseth legal because the -eth ± no not legal sorry ± religious because the -eth ± and the hereinafter ± I dunno might be legal could be I: Can you say why you think that? B: Why I think? sounds a bit ± a legal expression to me ± documents I don't really look at I must admit N: M: I: M: N:

Biblical They're like legal terms Why d'you say that? They're ± Well you could just say hereafter couldn't you rather than hereinafter ± hereinafter's a very legal term isn't it? I: What makes it a legal term? ± how would you know it's a legal term? N: Seeing it ± in legal things you know ± and witnesseth sounds very biblical 166

Folk Characteristics of Words

A number of informants, including N, M and B above, stressed the fact that such words occurred in the written medium: C: . . . they both strike me as a bit archaic I: Why what about them? C: Well you don't see the -eth ending anymore ± hereinafter is more pompous than . . . archaic . . . I: What makes it pompous? C: Because it's not in everyday writing not just not in everyday speech I mean it's not in in modern writing it you may get it in legal documents I don't know it's a mixture of pompous and kind of archaic of®cial D: Er ± well ± they're the kind of words you'd see in a legal document I: Why d'you say that? D: They have that kind of ± em ± antiquarian look and sound to them I: What d'you mean by look and sound? D: Well the ®rst one in particular ± em ± is the running together of three words ± em ± in a kind of formula that's very common in legal documents ± em ± yeah ± em and the second one ± well that's still used in some cases F: Well it's sort of like ± obviously to do with law and things well it strikes me to do with law and things there I: Why? F: `Cause of Old English I think I: How can you tell? F: Em ± by the way it's written as in one word and witnesseth is not a word that's used nowadays Q claiming that such words were `biblical' and `legal', very `formal and archaic', said that they were `words you'd only write'. W, after V had stated that hereinafter `shouldn't be joined up', replied `there's no such word . . . try and look it up in a dictionary'. The American law student H claimed to know the word hereinafter through acquaintance with his legal studies. He 167

Folk Characteristics of Words

commented that it was a term in the process of being eliminated and replaced by `more plain English': H: Hereinafter they use that see we have that word sometimes like in our legal stuff at school I guess I never did sit there and de®ne it though but ± it'd mean `from here on out' . . . I don't know for sure as a matter of fact in our classes they're trying to get us away from using big words they want us to speak more plain English When he discussed witnesseth, he said `sounds like er Old English or ol' King James Bible or something'. Amen and thee (qu.6) were also considered to be `religious' . . . thee is something like the word I was saying before like in King James' Bible'. Referring to amen and thee, subjects sometimes spoke of such words being `old' not in the sense of outdated, but rather in the sense of being olden and archaic: D: . . . words that er ± automatically a ± a kind of a religious application em ± one because it actually is religious ± er ± texts it's ± actually a very ancient word ± a Jewish word ± and the other because em ± it's a word that we associate with older translations of ± er ± especially canonical works ± canonical scripture G when referring to amen and thee gave a detailed (historical and synchronic) account of the words: G: . . . amen and thee ± well those are both seventeenth century terms aren't they, used in the authorised version of the Bible and still used in em em liturgical language today in the Church of England anyway and I think probably not far outside that but amen is still used in the Roman Catholic Church and in all the Churches I think but thee I think is probably now con®ned to the Church of England and has gone outside the realm of the Roman Catholic Church unless it's still being used by the Quakers 168

Folk Characteristics of Words

Again, as with hereinafter and witnesseth, these words were spoken of as being more frequently written than spoken: I: Em ± you said ± or you referred to them as being in texts (see above) D: Yeah I: Is that you only place where you ®nd them ± in written language ± do people use these words? D: Er ± probably the ®rst one amen a little more and the second ± em ± increasingly certainly when I've heard it it's used in a slangy way now And although M did not explicitly mention the fact that such words were more often written, she commented on an aspect of literacy on the use of the word thee N: M: N: I: N: M: N: M:

What have we got? ± some more biblical ones now Yeah very religious Yeah Why d'you say that? Well you say amen Amen comes at the end of every prayer And thee is quite often in a prayer And even more `biblical' with a capital `t'

However, the majority of my informants just commented on the fact that they were words associated with religion, God or the Bible: F: . . . well that strikes me as the Bible B: . . . well again it's religion strikes me on both counts E: . . . Biblical again innit really ± priest J: . . . well amen means `so be it' I: What sort of words are they? J: . . . well Church words or Biblical words Q: Religious terms . . . amen is a serious word . . . thee is very intimate, it's a very important word, to say it or not makes an awful lot of difference, nothing constrains us to say it. 169

Folk Characteristics of Words

A was unable to comment on the difference between this pair and witnesseth and hereinafter while K and L could not see any similarity between amen and thee and were only able to comment on the difference of meaning A: That looks like Biblical words again I: Can you say how you can tell? A: Well Amen is the ending of every prayer and thee is just how it's spoken in the Bible I: Do these words differ from witnesseth and hereinafter? You said they were Biblical words? A: They're just words I know appear in the Bible L: K: L: K: L: K: I: K: L: K:

Amen and thee are in the Bible aint they? Amen ± they don't even look together do they? Pardon They're not together words are they? Well they're like Bible words aren't they? Are they? ± cor you're talking a load of ol' mug you ± what are we supposed to be doing? you gotta what? You've got to compare or contrast `Well amen is an ending aint it? Yeah It's the end of something and thee that's like you aint it? yeah they're different

Scienti®c words Although the words of question 5, as with those of question 3, were unfamiliar to many of the informants, the former pair was not considered to be `old' or `archaic' by any subjects. Even if the informants had no prior acquaintance with the words in question, they often hazarded a guess at the subject area. `Science' was a common ®eld noted and the informants generally gave the `component' parts of the words as justi®cation for their opinions (see also p. 166 ff. above) 170

Folk Characteristics of Words

H: I don't know either of those words . . . . . . I don't think they're words that I've ever used knowing what they mean ± on purpose . . . ontogenetically that seems more scienti®c, I really have no idea what the ®rst one is I: Why does it seem scienti®c? H: -genetically sounds biological M: I: M: N: M: N: M: N:

Oh Lord ± haven't got a clue about those What sort of words do you think they are? Em Biology-type things Scienti®c: Scienti®c yeah Of some description `Cause anything that ends in -logically or -ically usually is a science

O: . . . I haven't got a clue what either of those mean I: What sort of words do you think they are ± what could you say about them? O: I would imagine they're both medical words . . . I: Why? O: Dunno just seem to me as that I dunno they're just onto-genetic- I presume it's got something to do with genetic, I could be wrong, being medical words I: Are they everyday words? O: No not as far as I'm concerned they're not no Other guesses were that epistemological was possibly a `religious' word (B) because of the word epistle, and W thought that both sounded `like words the Americans have made up . . . funny words'. Two informants based their judgements on having had prior acquaintance with such words through their reading: C: I: C: I:

They're sort of specialised words Anything else ± what sort of words Long words What makes you think they're specialised? 171

Folk Characteristics of Words

C: Well I know they are it's not that I think they are I know they are, I mean you'd see the only time you see epistemology ± ological well I see it in sort of history of ideas type things talking about thinkers and you know theory of knowledge that type of thing I mean you don't go round reading that in my paper The Sun ± and I imagine the same kind of thing with ontogenetically I mean I haven't seen that I: You've never seen it before? C: No I don't think so it could be a made-up word for all I know ± I know it could exist I: How? C: Because I've seen other onto- words G: . . . well they belong to philosophical language I think I: How do you know what makes you say that? G: Well just my reading of philosophical books . . . Q and D also considered one of the pair to be `philosophical': D: Well grammatically they're different I: What sort of words are they? D: Well grammatically one's an adjective and the other would seem to be an adverb ± em they're technical I: Why d'you say they're technical? D: Well you can see from their ± you can see from their etymology that they've been very carefully designed to er ± encapsulate a particular idea ± one philosophical and the other ± I think biological or medical Others claimed that such words were incomprehensible: `fucking big words' (X) `double dutch' (Y). E did not wish to make any comments at all about such words because, as he said, `I couldn't bloody read `em' E: Now I haven't got a clue I: Can you say anything at all about them? E: No 172

Folk Characteristics of Words

I: E: I: E: I: E:

What sort of words are they d'you think? What they are? Yeah Dunno ± I wouldn't even bother with `em myself personally Why? `Cause I couldn't bloody read `em ± I wouldn't even know what they meant I: Could you have a guess? E: No ± never in a million years ± I'd just look at that and go `well we'll miss that one out' P: Oo I don't know what those two words mean I: Can you say what sort of words they are? who do you think would use them or where they would be P: Obviously they're words where people study something at college or something y'know what I mean like I dunno I: Why do you say that? P: `Cause they're long The Layperson as Sociolinguist Although the pairs parsimonious and stingy (qu. 9) and old and mature (qu.12), were generally acknowledged by the informants as having essentially `the same meaning' or were identi®ed by the interviewer as having `more or less the same meaning', the difference was often treated in a different manner in the two cases: in question 9, it was often ascribed to the user of the word whereas in question 12 to the `word itself'. The judgements on question 12 generally did not concern the users of such terms. Two informants, S and G, explained the `linguistic' aspects of the words: S: To say somebody or someone is `mature' is an opinion ± you usually use it with the word seems ± old is a statement G: . . . mature is em a euphemism really I think so isn't it or it it's much more er complimentary and old is you know you say you're an old man you're a mature person em so I would 173

Folk Characteristics of Words

regard those as semantically rather different that old man has a very wide range of meanings er complimentary or uncomplimentary whereas mature is used in wine and that sort of thing And D explained how they could be `interchangeable' in some contexts but not in others: D: Old and mature ± they're closely relatable but they're not quite the same ± em ± em ± they they can be interchangeable in a lot of contexts but mature ± em ± can be applied in ± in er ± in contexts where it doesn't necessarily mean great age ± it can mean `old enough' whereas old ± old means great age Referring to old and mature, K considered them all `the same': K: Well they're the same as Jan [informant L] ± one's Jan ± old and mature they're both age aint they L: Yeah Although A claimed that they meant the same, he also judged mature to be `a kinder expression' and has a `meaning' not shared by old: A: . . . both mean the same thing ± old would ± could be used slightly derogatively whereas mature is a much kinder expression in our society and er ± mature could mean mature cheese or mature wine ± whereas old doesn't really say that These examples were brought up by N and M who also thought mature to be a `nicer term': N: M: N: M: N:

It's politer for women to say they're mature Yeah it's a nicer term Nicer term of old You don't say the cheese is really old ± you say it's mature Mature ± yeah you don't say old cheddar 174

Folk Characteristics of Words

M: Then again you wouldn't say great mature sissy ± you'd say old sissy E also thought that mature was a `nice way of saying old': E: Mature's just a nice way of saying old innit? I: Why is it a nice way? E: Doesn't sound so bad as old T, V and W also described mature as being `more polite' whereas Q claimed that it was `a kind word . . . more diplomatic, old doesn't beat about the bush it gets to the point'. One informant emphasised their semantic differences: F: . . . well they're obviously not really the same ± mature is unripe and unready and old is em old and decrepit I: So how would you know how they're different? ± what makes them different? F: Well one's old and knackered B brought up the issue of how the two words are often misused: B: Incorrectly used in a similar ± in a ± as being two and the same whereas they're not ± em old I would consider either to be something like em ± either em ± vintage or possibly senile or certainly beyond the word mature because mature can certainly be used for well people who're your and my age if you like you know Thus my informants above, in speaking of the pair old and mature, generally referred to their differences of meaning. This contrasted with the pair stingy and parsimonious which evoked a response based upon, not the `words themselves' but upon the imagined type of users of such words. E: Stingy's tight ± don't know about that one I: Well it means the same ± so would would you say about it now? what sort of word is it . . . What's the difference between parsimonious and stingy? 175

Folk Characteristics of Words

E: One's for the upper class and one's for the lower class probably J: Well stingy is another is a word you'd use instead of for a mean person or for somebody and parsimonious is is the same sort of thing only stingy is more commonly used more commonly used I: Who uses the other one do you think? J: More of an educated person I should imagine B: I don't know what the ®rst one means at all I: They mean about the same B: Well if they mean the same well that's it ± then I suppose one is a correct word whereas that one really is a slang word I: What would the difference be then ± just slang? B: Yeah I: But which word is used more often? B: Oh ± de®nitely the second: I: And what sort of person would say the ®rst? B: I don't really know ± I don't think I've ever met one ± hence I don't know the meaning of it ± I think if somebody had said it to me I would run off to a dictionary and try to ®nd out what he was on about ± em parsimonious let me see ± possibly a high ranking cleric ± judge possibly ± professor at college ± university ± especially if he's trying to be clever Others claimed that the distinction was that between the written and the spoken word, that stingy was a more `common word': stingy `is a more colloquial word, you'd use it more often' (R). M and N drew a distinction between `your average bloke' and `pompous people' and added that parsimonious was found more often `in books': N: What's parsimonious mean? I've forgotten I: It means more or less the same as stingy N: The same as stingy ± that's what I thought I: What's the difference between them? 176

Folk Characteristics of Words

M: It's just the way you talk really I mean your average bloke off the street will say you're stingy and ± N: Yeah but you don't ®nd your average bloke on the street saying you're parsimonious I: Who would say that? N: You'd get it in books wouldn't you? M: Yeah it's literature em N: It's not a word you get in everyday conversation M: People ± pompous people use it N: Yeah pompous and parsimonious as well C thought that parsimonious was a word which one would ®nd in an `academic book' but that stingy was a `nicer-sounding' word: C: Parsimonious and stingy well they don't mean the same thing em ± number one is the kind of thing you read in an academic book and number two is the sort of thing you wouldn't read in an academic book er stingy is a far better word I: What do you mean a better word? C: I think it em ± parsimonious like a lot of those sort of longer Latin words it's not very colourful there's a kind of ± whereas stingy sounds right someone who's a bit mean, I'm just talking about the sound of a word I mean you get a lot of words like harmonious I mean parsimonious sounds a lot like other ones Apart from referring to the `literary' aspect of parsimonious, G emphasised that stingy was a more `emotive' and `nasty' word: G: I: G: I: G:

. . . mean exactly the same to me What's the difference between the two words then? I don't know then . . . They're absolutely synonymous you're saying? Em not exactly stingy has er an emotive sense I think whereas parsimonious doesn't really, I would say parsimonious used in literary sense em ± meant that someone was very careful with his money but stingy meant that someone was nasty . . . 177

Folk Characteristics of Words

I: So the emotive sense is nasty with stingy that's the difference? G: Yes I think so yeah stingy stingy is an emotive word I would think parsimonious is a a er more literary . . . a more literary em ± nasty word at least that's that's the way I would see it I: What do you mean by nasty word? G: Well a word in a way of criticising somebody, I think parsimonious is a word where you criticising somebody ± em but ± em you're doing it in a rather er more civilised way than saying stingy where you're saying it in a much more you know forward way A did not think that anyone would say parsimonious: A: That one ± the shorter of the two stingy ± is more common term in our society's language but you wouldn't very often hear that said and you wouldn't very often see it written I would think I: What sort of person do you think would say it? A: Parsimonious? I: Mm A: I don't think anybody would whereas Q, despite ®nding both words unpleasant, thought parsimonious a more educated, literary word: Q: . . . nasty words no nice connotations like saying `Mr Bleeny' which is a horrible word, `a' is a more educated and literary word `b' is a bit nastier Spoken and Written Words The view of some of my informants that parsimonious, amen, hereinafter and alas were words which were more frequently found in the written medium (p. 167 above), contrasted with the opinion that certain words, such as ouch and hey (qu. 2) and gazooks and snortle (qu.11), occurred primarily in the spoken medium. However, in both cases, even if they never had any prior 178

Folk Characteristics of Words

acquaintance with the words in question, they were able to speculate on their meaning or use. At ®rst C, because he had never seen the words of question 11 written before, was unable to make any comment. However, when asked to make a guess about gazooks, thought that perhaps it did `ring a vague bell'. Likewise O, who had only ever heard gazooks but `hadn't a clue' about what it meant, was nevertheless able to guess as to its meaning. C: I've never seen gazooks written before so ± snortle ± I don't know what snortle is ± so big question mark is my immediate reaction to those two I: And nothing at all strikes you odd about them C: Well except that they're odd ± gazooks I: why're they odd? C: Well I've just told you I've never seen them written before ± snortle em I: What sort of words would they be if you knew what they meant? C: I think gazooks is something along the lines of er ± gee whizz em ± I don't think anyone would say it nowadays ± I think it's a word you would say [sic] I: Is that because you've heard of it? C: Gazooks yeah it rings a vague bell O, although he too had heard the word before, thought it meant `shock': O: gazooks I've heard that word but I haven't got a clue I think it means shock I: You think it means `shock'? O: Yeah instead of walkin' round sayin' bloody hell, gazooks that sort of word I: Why do you think that? O: Well it just strikes me as that type of word The words above were labelled by six informants as exclamations which, according to W, `you would say if someone hurt you'. He 179

Folk Characteristics of Words

did not consider them to be `proper words' because `you wouldn't ®nd them in a dictionary'. M, de®ning an exclamation as `a re¯ex sound' claimed that such sounds were meaningful: M: I: M: N: I: M: N: M: N: M:

I'd say exclamations What's an exclamation Sort of re¯ex sound ± somebody thumps you You go ouch ± and if you receive something you go hey Does it ± do they mean anything? Em Everyone knows what they mean don't they? Hey is different that one you know has got different meanings Yeah Whereas hey you could be calling somebody ± somebody bumps into you you say hey ± you know

For G etymology was a clue to the lexical identity of gazooks: I: G: I: G: I: G:

What type of words would you say these two are? Do you mean grammatically? Not necessarily well if you want to say grammatically Well gazooks I would have said was an exclamation Can you say why? Er er only historically because it means God something or other ultimately so it must be, er em how it's used at the moment I don't know but historically it's got to be an exclamation

N and M thought that the words of question 11 were `crossbreeds' between words and also proffered an explanation for the etymology of gazooks: M: They sound like em I: It's like when you say `gad-zooks' M: Like snort and em chortle or something like that and snorkle N: They sound like one of those em ± what d'you ± you know M: Typical Jabberwocky 180

Folk Characteristics of Words

N: Countri®ed versions d'you know what I mean? ± when they countrify them I: Why do they sound like that? M: Probably because we've never heard of them, they look very similar to words ± you know N: Well snortle is de®nitely a cross between two words ± well you know you could imagine a snortle can't you ± it's sort of a an annoyed snortle I: What about gazooks? M: It sounds like a shortened version of gad-zooks N: Gad-zooks yeah Ouch and hey, apart from being considered `slang' (see p. 161 above), also elicited comments on the sound/meaning correlation; D used the term `ejaculation' to express this point: D: I: D: I: D: I: D: I: D:

. . . similar ± similar words Ouch and hey? Yeah ± em What sort of words are they? Em ± they have the ± er force of ejaculations so that's really what they are What do you mean by that? Em ± kind of ± em conventionalised ± em ± words to convey sounds made by people Are they words? Yes ± yes they are words

Some informants either did not think that gazooks and snortle were words, or considered them `made-up' words from children's literature, mentioning children's writing, comics and nursery rhymes: `comic words' (R), `from Batman' (V), `from children's comics' (Y). A came to this conclusion after ®rst offering the opinion that they were not real words (as did X `made-up words'): A: . . . they don't seem like words ± they seem like words that I: Why not 181

Folk Characteristics of Words

A: I: A: I: A: I: A: I: A: I: A:

someone's just made up Why don't they seem like words? I've just never seen them before But you didn't say that of ± er about epistemological No that's true -m yes ± I do ± they just don't sound like words Why ± what in them doesn't sound like a word? . . . gazook I mean I don't know a word that is zook and I don't know a word that's is snortle But you know a word snort Snort that's true ± yes ± I do ± I wonder why that is? Can you imagine any places in which these words might occur? Yes in ± in a nursery rhyme or ± em a story on television ± a children's story

And two informants, T and P, were unable to pass any judgement on these words: P: I: P: I: P: I: P:

Well snortle's one thing you breathe innit No that's snorkel Oh, well what's snortle then? I dunno what gazooks are (sic) Can you think of what sort of word it might be? Gazook? Gazooks or snortle? No

These words, like ouch and hey elicited comments on the sound. For D, H and J, snortle reminded them of a noise made by the nose: D: Well it probably comes from snort ± er ± I don't know ± perhaps it's some kind of ± yeah perhaps it represents the noise made by your nose in accordance with some particular activity ± I don't know H: snortle I guess is laughing through your nose J: . . . snortle is is I should imagine to make a noise through the nose 182

Folk Characteristics of Words

I: J: I: J:

What sort of word is it? What do you mean what sort of word is it? Is it a common word? Reasonably so yes

O was reminded of a `dirty laugh' made by an old Englishman: O: . . . snortle again that's a sort of dirty laugh I think er you want me to say what kind of person I: Yeah O: Old Englishman I'da thought yeah Three informants used the term `expressions' to refer to ouch and hey which, according to G, was something requiring an exclamation mark G: Ouch and hey em well two em what d'you call it ± expressions aren't they I think em one which means you've been hurt ± hey I don't know ± perhaps saying hello to somebody I: So they're just expressions? G: As far as I'm concerned yes yes er ± I: What do you mean by an expression? G: Well something that needs an exclamation mark after it I: Well what sort of words are expressions then? G: Em words which you use when you express something er suddenly, I think I would say ouch when you're hurt hey or hi whatever it is when you meet somebody or something like that I: Do they have a meaning? G: Em well yes em one is em an expression of some sort of rather mild displeasure and the other one is an expression of some mild pleasure I would have thought B thought that gazooks was an expression similar to ouch I: Why is gazooks an expression? B: Why is gazooks an expression? ± I've only ever heard it being a `O what gazooks' you know 183

Folk Characteristics of Words

I: How does an expression like gazooks differ from slang or swearing? B: Well it it ± I suppose gazooks is really a ± it's it's a it's an expression really like ouch okay? And H termed gazooks `an expression of surprise'. For some of the informants, the words discussed above were meaningful. For H, ouch and hey had a similar meaning. H: Er ± I'd say they're similar in that they're both words that we use off the top of our head I: Do they have a meaning? H: Do they have a similar meaning? I: Do they have a meaning? I mean you said they're words that you say off off the the top of your head H: Hm ouch does I guess but em I don't know for sure what hey would would mean other than trying to catch someone's attention and they're ± I: But you think they are a bit similar? H: Well yeah I think they're kinda used in the same way sometimes whereas for E they were dissimilar words with different meanings: I: Anything to say about whether they're similar or ± E: No they're not similar are they? I: What's different about them? E: They sound different ± they're different things I: What do you mean? E: Hey is like ± what d'you say about hey ± hey man I: What sort of word is it? E: Hippie word innit I: And ouch? E: If you burn yourself innit ± pain ± ouch K and L only discussed the `meaning' of such words: 184

Folk Characteristics of Words

K: Ouch is when something hurts aint it or ± L: Yeah hey is when you greet someone K: What cows eat L: No that's h-a-y I: What sort of words are they? K: One's pleasure and one's pain I expect Cuckoo and crisp (qu. 8) were included partly so as to see whether the conventional `sound-meaning' relation of the ®rst word would evoke any (unconventional) sound-symbolic nature of the second, partly to investigate how subjects would deal with pairs of words having no immediately recognisable dimensions of similarity. Only seven of my informants commented on the onomatopoeic aspect of the words and, of the seven, all but one were interviewed with a partner. The two pair V and W and X and Y said they were `written as they sound' and S said `They sound as they are said'. M knew that there existed a `word' for this phenomenon but could not recall the term: M: . . . those are words that when you say them they sound like ± like when you bite a crisp N: It sounds like it crisp yeah crisp's the noise a crisp makes M: I've forgotten the word for it now D, although he commented on the `representational' nature of the ®rst word, was unable to see any similarity with the second: D: . . . Em well cuckoo's a bird and the word's an interesting word in that it actually attempts to represent ± the ± er call of the cuckoo as it were ± er crisp can be a noun or an adjective I: Hm mm ± no similarity at all? D: I wouldn't have thought so ± no The majority of the informants commented on the meanings of the words in the question and, even when asked for further or additional comments, were unable to `see' anything else 185

Folk Characteristics of Words

B: . . . oh what? ± compare or even contrast ± well there's no comparison ± and there's no way of really contrasting them they're so different ± a cuckoo is a bird ± crisp is potato J: Well well I can't see any connection between the two whatsoever cuckoo is either a bird or er ± in another form could be not quite right in the head and crisp is ± crisp ± it's ®rm crunchy er brittle O: . . . cuckoo I immediately think of cuckoo clocks er crisp crisp one pound note, they're both words as far as I'm concerned H: . . . okay I don't see any relationship between them cuckoo that means weird or crazy ± er well you got the cuckoo clock with the funny lookin' bird but I don't see how that relates to crisp even when asked to `disregard the meaning' G: . . . mmm strange bedfellows ± I'm not sure that I understand why those two are together ± are they supposed to be adjectives? I: Well I'm not saying, it's up to you to say whatever . . . G: They don't seem to me to go together at all em ± but if they are adjectives I would suspect they meant something like strange or ± odd or something I just don't know about those two I'm afraid I: Okay G: Cuckoo means strange to me or peculiar or different or someone who's a bit peculiar crisp as an adjective doesn't mean anything to me at all I: Well disregard the meaning for a moment well apart from the meaning of the word ± think about them again . . . G: . . . Well I mean a cuckoo is a bird and a crisp is a thing you eat out of a packet I:

Yeah yes well you don't have to explain what they mean just look at the word G: Oh I see well that's what they mean to me all right 186

Folk Characteristics of Words

Although these words begin with `the same letter', units smaller than the suf®x -eth and pre®x -onto (cf. p. 171f. above) were not as readily identi®ed by the informants as worthy of comparison. In fact only two informants commented (in passing) that these words began with the same letter but did not pursue the point: A: Cuckoo and crisp em ± both begin with `c' ± cuckoo's a bird and crisp can either be a potato crisp or crisp sheets I: You can't see how ± why I've put those two words together at all? A: No not really E: Well cuckoo's a bird innit? I: Yeah E: And crisp's a crisp ± a little bit of potato crisp ± been cut up and ®red in all different ¯avours I: Can you see why they may have been put together ± cuckoo and crisp? E: Start with the same letter? Informant R however, saw a similarity between these words claiming that both were `descriptive words of spring'. In fact, the metalinguistic labels descriptive, descriptions, and describing were those most frequently applied to all the words under investigation. However, such labels have often been used as grounds for philosophical debates.

Descriptions Apart from calling particular words `exclamations', `swearwords', and `expressions' and identifying them as `legal', `pompous', `common', or `old' a large number of informants spoke of many words of Question No. 3 as `descriptive' or `descriptions', or as `describing' in particular or in general. What such individuals meant by such words or what criteria were being employed was far from obvious. For example, referring to 187

Folk Characteristics of Words

question 1 (antidisestablishmentarianism) and (it), B spoke of such words as not being `descriptive' whereas D thought of them as `describing': B: Well ± I should imagine that they're both in some ways very similar but totally different in other ways ± the similarity I can see is being em ± antidisest- whatever is em ± again it's not really descriptive ± and it is not descriptive either ± so that's why would see the similarity D: Functionally similar in that they're both describing or can be describing abstract things ± em ± just one happens to be a very long word ± in fact the ®rst word can probably be replaced by the second word in a great number of contexts Referring to question 3 (hereinafter and witnesseth), X only commented that they were `descriptive words' and for S, epistemological and ontogenetically were `scienti®cally descriptive words'. In such cases these informants did not state what such words were `descriptive of'. F however, when asked, said that gazooks and snortle (qu. 11) were `descriptive' `of an emotion': F: I: F: I: F: I: F: I: F: I: F:

. . . gazooks a good word A good word? Yes What do you mean by a good word? I love it it's a good word What makes it good? It's descriptive and a nice word gazooks . . . Why do you think they're good words? Because they're descriptive Descriptive? Of what? Em of an emotion you want to say something that's em not necessary but to make people sit up and listen

For R old and mature (qu. 12) `both can describe wine' whereas for T `they can mean the same thing or be used to describe something different, `mature' is more polite. E thought that shit 188

Folk Characteristics of Words

(qu. 7) was not a swearword, but was rather `describing something' as was parsimonious (qu. 9): I: E: I: E: I: E: I: E: E: I:

Well you said it's not a swearword It's not a swearword is it ± it's not like cunt or wanker is it Are those swearwords? Yes . . . What sort of word is shit if it's not a swearword? Describing something What does it describe? Turds Stingy's tight ± don't know about that one Well it means the same ± so what would you say about it now? ± what sort of word is it? E: It describes something innit?

C also thought that certain swearwords are more `shocking' than others because they describe certain activities: C: Well I mean that the word fuck describes the act of sex and you'd tend to ®nd that em ± you know the stronger ones the ones you know to go up from fuck you get a kind of a specialisation of it like motherfucking and things like that er I mean that obviously describes the a slightly more shocking thing em ± I don't know what blimey refers to ± though the exception to that rule would be bugger which isn't as strong but which describes something you know Some informants also spoke of some of the questions in other tests as `describing'. For example, when asked to comment on the underlined words in Question No. 4, R said of John (qu. No. 1) `it is describing John in a derogatory way', of dead (No. 2) `it just is describing' and of mad (No. 3) `describing again'. V referring to the `meaning' of it, contradicted W's claim that it `doesn't mean anything on its own', with the statement `Yes it does, it's a descriptive word'. This chapter has illustrated that `subjective' terms, such as good, bad, nice, pompous, descriptive are salient in the 189

Folk Characteristics of Words

metalinguistic vocabulary of language-users. But these terms, as mentioned on p. 147 above, are of no interest to the professional linguist. They wish neither to accept nor refute such value judgements. Crystal, for instance, says that the linguist `is not trying to evaluate the language in terms of some aesthetic, moral, or other critical standard' (Crystal 1977: 8); rather he tries to be `as objective as possible, and aims to avoid the misconceptions about the nature of language and languages which have been so dominant' (ibid.: 26). And it is this discrepancy between theoretical pronouncements and lay opinions that give rise to the sort of linguistic hypocrisy as identi®ed by Amis in his review of Weiner's The Oxford Guide to English Usage: the compiler of the present volume remarks that what interests him is `the degree of acceptability in standard English of a particular use rather` than any `dogmatic' question of right and wrong, but after thus touching his hat to linguistic egalitarianism he settles briskly down to telling us what to do and what not to do, and quite right (Kingsley Amis, The Observer, February 19 1984). What is observed by Amis is that both prescriptivism and antiprescriptivism (descriptivism is the professional linguists' favoured term) are part and parcel of the same activity: attempting to describe or de®ne a language. As Cameron argues Both prescriptivism and anti-prescriptivism invoke certain norms and circulate particular notions about how language ought to work. Of course, the norms are different (and in the case of linguistics, they are often covert). But both sets feed into the more general arguments that in¯uence everyday ideas about language. On that level, `description' and `prescription' turn out to be aspects of a single (and normative) activity: a struggle to control language by de®ning its nature (Cameron 1995: 8). It is to a discussion of a major in¯uence for our normative activities that I turn to in the ®nal chapter. 190

7 Reorientation: the Integration of Speech and Writing

So what conclusions are we supposed to draw about this unit called `the word'? It has been shown both by me and by my informants to be at once an illusion, an invention of grammarians, an artefact of traditional wisdom, of literacy and of pedagogy. In this concluding chapter I will attempt to spell out how an integrationist would answer the question `What is a word?'. The Integration of Oral and Literate Skills This study has demonstrated that there is much diversity as to what native speakers of English understand by the word word. It would appear that the use of this term is varied, and depends, among other things, on the `literacy' level of the individual. In addition, there does not seem to be any way of systematising my informants' comments. Rather, the very variability of the judgements highlights both the indeterminacy involved in everyday communicative acts and also the failings of modern linguistic theorising in its attempts to formulate a `correct' theory of language. The survey has thus set out to demonstrate, in part, that there is no `right' or `wrong' about the question of wordhood. Accordingly, my approach supported the claim that language can be appropriately investigated only by examining its use by ordinary, i,e, not ideal, speakers. It also took as its object an individual response to an understanding of linguistic experience, i.e. any event which is taken to be one of using 191

Reorientation: the Integration of Speech and Writing

language. Both that experience and its understanding will depend on the speci®c cultural contexts into which they are integrated. Since different individuals and the same individual will be attempting to achieve a particular goal at a particular time in a particular context, there is no question of the `words' they use being invariant linguistic units. Any question as to what a `word' is will therefore presuppose certain cultural practices, but the answer will depend on the `language game' into which the question is integrated. Although there is little conformity in individual metalinguistic responses, there are a number of researchable, cultural determinants. Within linguistics, only an integrationist approach can deal in a satisfactory manner with problems such as those of the recognition of any linguistic units, in particular because other theoretical frameworks have no systematic way of accommodating the key anthropological distinction between literate and preliterate cultures and thus coming to terms with the fact that literacy rede®nes language (Harris 1983, 1986). Literacy makes a radical difference to the way people think about themselves and their own social activities including language. In one sense, this book is an exploration of one aspect of the `literacy hypothesis' (Olson, Torrance & Hildyard 1985, Olson and Torrance 1991). This hypothesis does not only identify literacy with scribal competence, but rather construes it in a broader sense, `bearing on the general competence required to participate in a literate tradition' (Olson 1991: 252). This competence is re¯ected in the use that individuals, whether they can read or write or not, make of metalinguistic vocabulary. There is no reason to assume that the doctrine of the primacy of speech1 is valid for the linguistic perceptions of a literate society. Both the views of the linguistic orthodoxy and my informants are a product of literacy and consequently add to a continuum of diversity rather than to any con¯ict of evidence. Their views too have to be incorporated within the spectrum envisaged by the `literacy hypothesis' (p. 192 above). 1 `speech is prior to writing not only historically but also genetically and logically' (Bolinger and Sears 1981: 274).

192

Reorientation: the Integration of Speech and Writing

The problems that segregational linguists get themselves into over the `word' (p. 7 ff. above) stem in part from their attempt to generalise from a culture-speci®c term, in part from their strict imposition of a bi-planar format on a multi-faceted and openended concept. A sign is a sign only by virtue of its communicational context, and anything in context can acquire linguistic relevance, as is highlighted in the responses of my informants. One therefore cannot delimit in advance the features that will turn out to be communicationally signi®cant. Instead, directly asking speakers questions about wordhood provides `a means whereby necessary forms of social and communicative knowledge (imagined, stereotypical or real), can be brought to the surface and articulated in a form that becomes researchable' (McGregor 1990: 107). It is presumed that such knowledge `could only have been deduced or based on guesses from necessary forms of social and linguistic knowledge, whether this knowledge was based on real or imagined interactional experience' (ibid.: 108). The experience relevant to this book is that of participation in a literate tradition. Once communities change from being pre-literate to literate, `speech becomes integrated with a biomechanically different mode of communication, which depends on visual rather than auditory signs' (Harris 1998: 226). David Olson argued that the distinction between an oral and literate society does not simply re¯ect a difference between the spoken and the written word (Olson 1991). Rather, a literate society is characterised in part by the metalanguage it employs to refer to linguistic activities: Literacy involves the knowledge that language exists as an artifact, has a structure, is composed of grammatical units including words and sentences, has a meaning somewhat independent of the meaning intended by the speaker, and, perhaps most importantly, that its structures may be referred to by means of a metalanguage. In short, it is the belief that language can be treated opaquely, as a structure in its own right (Olson 1984: 223). 193

Reorientation: the Integration of Speech and Writing

In those cultures where language is considered `natural', part of one's biological make up for example, children learn but are not taught to talk precisely because language is not seen as a formal autonomous structure. There are no `naming' games associating words with things ± practices in Western society that even nonliterate parents engage in (Heath 1982). The assumption that language exists as an artifact is not necessarily expressed in or learned through reading and writing. Rather, it is the exposure to an oral metalanguage, which is the consequence of literacy. Olson refers to `The mental skills hypothesis' (Olson 1991: 257) which `allows that skills transfer across media and modalities'. The knowledge one aquires through reading and writing may also be applied to oral and auditory activities. Thus the knowledge acquired in reading and writing may apply in speaking and listening . . . Having learned to recognise words in written texts, one may come to recognise the words present in one's speech. And having learned to read a text and analyse its implications, one may come to treat an utterance as if it were a `text' and subject it to a similar sort of analysis, and so on (ibid.). Since this knowledge may be communicated orally, `the directness of the link between actual reading practice and the acquisition of literacy skills is indirect indeed' (ibid.: 258). It is not simply that children are applying the skills acquired in one domain to another domain so much as that they have acquired some knowledge about language . . . knowledge that is peculiar to particular literate traditions (Olson 1991: 258). Literate parents often teach their children `an orientation to language which will not necessarily be relevant to learning to speak but which will be relevant to learning to read' (Olson 1984: 226) Such observations call for a re-examination of what Saussure said about the interference of writing with people's conceptualisation of speech (Saussure 1922: 45). He claimed that although 194

Reorientation: the Integration of Speech and Writing

a language and its written form constitute two separate systems of signs and that the sole reason for the existence of the latter is to represent the former (ibid.: (p. 5 above)), the native speaker frequently con¯ates the two systems. Not only is the languageuser apt to assign as much if not more importance to the written sign, but the written form ± especially in highly literate communities ± `may give rise to erroneous pronunciations' (ibid.). These `distortions' are, according to Saussure, purely external, caused by social conditions: `Linguistics should keep them under observation in a special compartment: they are cases of abnormal development' (ibid.: 54). (He would say, presumably, that this is just what the informants' evidence shows.) But the interesting question this research has raised is to what extent the empirical evidence does or does not support a different hypothesis: namely, that literacy leads people to conceive of words as linguistic units that happen to have two (imperfectly) correlated forms ± spoken and written ± rather than as separate units that belong to different modes of communication (see in particular chapter 3 above). The linguistic sign, as de®ned in integrationist terms, presupposes a communication situation and is regarded as a communicationally functional unit. Language, from an integrationist point of view, is constituted by the integration of various communicationally relevant forms of activity, both vocal and non-vocal; and in the case of literate societies this normally entails the communicational integration of speech and writing (Goody 1968). The most obvious advantage of treating the linguistic sign as an integrational unit de®ned by communicationally relevant practices of communities and individuals is that it avoids the theoretical impasse discussed above (Ch 1), which ensues from basing linguistic analysis upon invariants of form and meaning. Instead, an integrationist's approach to the problem of linguistic units pursues to their logical conclusion the implications of the Saussurean dictum that in linguistics it is the viewpoint adopted which creates the object (Saussure 1922: 23). It seems reasonable to expect to ®nd in a literate society that the relevant integrational factors in the recognition of linguistic 195

Reorientation: the Integration of Speech and Writing

units will turn out to be related to literacy and degrees of literacy ± i.e. to re¯ect an individual's educational background, occupation etc. This research has shown that the lay speaker's linguistic/ grammatical intuitions, which segregationists have for decades treated as basic evidence for the construction of lexical and grammatical analyses, are a consequence of being acquainted with literacy at various levels. Thus there is a strong argument for the view that the integration of speech and writing must be a ®rst principle for any kind of linguistic analysis. Conceptual pro®le of the word in contemporary English lay metalinguistics There have been a number of studies of speakers' metalinguistic awareness (e.g. Barton (1985), Berthoud-Papandropoulou (1978), Cazden (1972), Hakes (1980), Piaget (1929), Papandropoulou and Sinclair (1974), Tunmer, Pratt and Herriman (1984)), but the concentration has been on the metalinguistic awareness of children, and the focus on a comparison between an adult's and child's conception of wordhood. The purpose of such psycholinguistic research has been to show the cultural and cognitive factors which in¯uence the development of such an ability (e.g. Levelt, Sinclair and Jarvella (1978)), in order to present a more detailed account of the `stages' by which a child eventually acquires an adult's level of metalinguistic competence. This study has differed from the above investigations in one fundamental way: it has rejected the assumption that there is a qualitative difference between a child's and adult's comprehension of the term word. It has also rejected the notion of singling out `a typical group' of language-users to provide a standard of metalinguistic awareness of words. This objection would apply to Chomsky's `ideal-speaker-hearer'. Although Chomsky does address the question of the individual's understanding of wellformedness of sentences (Chomsky 1965: 18ff), he then proceeds to put his individual into a hermetically-sealed laboratory. Instead, the empirical evidence presented here has supported 196

Reorientation: the Integration of Speech and Writing

the integrationist thesis concerning the indeterminacy of the linguistic sign. The results have stressed the integration of various contextual in¯uences, activities and abilities on an individual's understanding of the term word This contextual integration necessarily, and non-trivially, includes the individual's interpretation of the point of my question (see p. 50 above). The metalinguistic repertoire used by my informants was very open-ended and unpredictable. Sometimes a technical metalanguage was used to characterise a particular word under investigation. Informants D and J, for example, used phonetic to refer to the spellings in Questions Nos. 8 and 9 (pp. 71±72 above). And all my informants except for O, P and E used the words consonant, vowel, grammar, nouns, adjectives, and verbs, and even these latter informants were familiar with some of these words. O: grammatically I can't remember what grammatically means, I was good at English at school . . . Adjective is a describing word, that's about all I can remember, I can remember the names but I can't remember what they mean I: Do you know what the letters a-e-i-o-u are called? E: Nouns or verbs . . . what then? ± this is like going back to school I: Vowels . . . have your heard of vowels? E: Yeah I: Do you know what the other ones are called? E: Nouns I: No consonants E: Consonants ± yeah ± I watched that programme once ± I should know this. However, even those informants who did use such words often employed them in an idiosyncratic way. H, for instance, spoke about the improper grammar of the spelling l-u-v (p. 75 above), and A thought that love in I love dancing was an adjective. A large number of my informants often appeared to use word to refer to a formal, orthographic, non-acronymic, unit of two or 197

Reorientation: the Integration of Speech and Writing

more letters, lending support to the segregational view mentioned earlier that a language is a structure whose meaning is speakerindependent. O: I's `not a word is it, it's a letter' (p. 57 above) E: `I wouldn't say a was a word' (p. 58 above) . . . `TV aint a word is it? . . . abbreviation for a word' (p. 60 above) . . . `it's just a sound . . . you don't look at a book and start reading em em do you?' M: `those aren't words, those are numbers' (p. 58 above) P: `ABC aren't words they're letters' (p. 59 above) J: radar is `not a word . . . though its used it's not actually a word . . . capital letters mean `radio' and something' (p. 59 above). H: em `it's not for formal writing but it is I mean people talk like that' (p. 64 above). A: `pervert and pervert they're different pronunciation and they're still the same word but they mean different things as well' (p. 50 above). K: `the words are the same but the meanings aren't' (p. 52 above) B: `Well if you like they're the same words but they're being used in a totally different way' (p. 54 above). These views strongly contrasted with G's (the linguist's) opinions on wordhood, which clearly was not dependent on orthography: `hide and seek's got to be one word it's a it's a one thing'. (see also D's comment (p. 65 above) that a word `is any one of a number of different units used orally by human beings to convey mental ideas'.) For G, the identity of a consonant was purely dependent on the medium in which it occurred (p. 78 above), whereas for the other informants it was seen to be a written unit. These results seem to bear out Hamilton and Barton's (1983) ®ndings that those informants with a high level of literacy often referred to the spoken functions of the word word, while those 198

Reorientation: the Integration of Speech and Writing

with a low level of literacy frequently referred to the written aspect of this unit. Hamilton and Barton conclude that the use of the metalinguistic term word `was frequently idiosyncratic' and that `people gave the impression that they had picked up isolated pieces of information and terminology as they progressed through school' (ibid.: 590). the properties of word that will be salient to people are those which relate most strongly to the activities in which words are used with awareness. In this way, adults and children in the process of learning to read and write may have more in common with each other than most people have with linguists, or than people (of all ages) who have experienced formal schooling have with those who are unschooled (ibid.: 593). In fact, many informants referred to schooling as the source of their metalinguistic judgements: B: that's what I was taught [don't is two words] (p. 61 above) M: Again that's how I was taught ([the spelling of realize] p. 73 above) H: . . . we always learned in school that aint is not a word is improper English and that we shouldn't use it in formal writing (p. 137 above) J: Well grammatically it's not correct . . . my English teacher told me so a million years ago (p. 143 above) O: they're not words that you get taught at school if you know what I mean [blimey and shit] (p. 63 above). Informants also referred to the signi®cance of certain graphic conventions and diacritics. For example, M referred to the use of an upper case letter on a word's connotation, claiming that thee `is even more Biblical with a capital t' and N, her interview partner, referred to the effect of punctuation on a word's identity 199

Reorientation: the Integration of Speech and Writing

M: N: M: N:

WRITTEN written it's only one, spoken it's four Why is it only one written? Because those aren't words those are numbers Yeah but if they were written as words it would be four wouldn't it? It all depends how you look at it M: You call them letters and ®gures N: But only because they've got commas after them they're each a separate thing And both C and G, referring to ouch and hey, stated that the words required an exclamation mark when written. G said that they were both `expressions' `something that needs an exclamation mark after it . . . words which you would use when you express something er suddenly' (p. 183 above), and C said that `they're exclamations you would rarely see it without an exclamation mark after either of them'. G was also of the opinion that the `expletives' blimey and shit `need an exclamation mark after them'. Olson and Astington (1990) and Torrance and Olson (1987) found that although school-based activities, such as reading and writing, are necessary for a degree of metalinguistic competence, by far the greatest in¯uence is participation in `a certain form of discourse' (Olson and Astington 1990: 711). Talk about texts is just as important as the skills of reading and writing. My informants made reference to a large number of texts from television programmes (O: you only gotta look in the TV that's how they spell it without the `u' [colour] (p. 74), advertisements (p. 65), computer software (p. 75), and academic books (chapter 8) to dictionaries. The dictionary, as might have been expected, appeared to play a key role in informants' understanding of what is meant by the word word. And since the dictionary is an outgrowth of literacy, standardisation and printing, and used in early pedagogy, one would expect it to be in¯uential in the shaping of one's metalinguistic awareness (see Davis 1999). The dictionary was also appealed to by my informants to justify their views on linguistic correctness and `proper' English (see pp. 62±63 ff.). Some informants, even if they disagreed with or did not understand certain lexicographical de®nitions, refused to accept that the dictionary was `wrong': 200

Reorientation: the Integration of Speech and Writing

A: I would term it not to be a word or think it not a word if it wasn't in the dictionary . . . I: So you disagree here the way they've de®ned cat? A: Yes I do I: So you're saying that dictionaries can be wrong? A: It's not wrong ± that de®nition is not wrong ± but it's ambiguous O: . . . I'd have to agree with what it said `cause I don't know any different I: So you think the dictionary is usually right about how to de®ne something? O: Oh yeah that's right for the words that I know yeah In fact the concepts `correct', `incorrect', `right' and `wrong' played a large role in the metalinguistic judgements of my informants. Such concepts were often used `conventionally' to refer both to expressions which were not `grammatically correct' and to the spelling of certain words as being `correct' or `wrong'. However, they were also used in an idiosyncratic way: the metaphorical expressions in Question No. 4 were termed `incorrect' by both J and W (pp. 112±113), whereas V applied `incorrect' to em. In neither case was it made clear what the actual nature of the incorrectness was: all three informants were able to articulate how such `expressions' were `usually' used. This was paralleled in the statements that words were `real', `good' or `bad'. In such cases it was unclear what the criteria were for such judgements. For example F, referring to the two spellings of realize, said `they're both quite good words'. H declared aint to be always `a bad word' and that `it was not a word but we could still ®nd it in the dictionary' (p. 137 above). And stingy was said by P to be not `the proper word'. The concepts of `right' and `wrong' etc. are probably salient in the metalinguistic repertoire of laypeople because they can be applied to all forms of social behaviour: as de Beaugrande writes `the question of how people know what is going on in a text is a special case of the question of how people know what is going on in the world at all' (de Beaugrande 1980). 201

Reorientation: the Integration of Speech and Writing

I often got the impression in some of these cases that the less literate informants did not have any strong or de®nite views about the questions but were just producing answers they thought I wished to hear (see E's comment `I'm just trying to be clever here and it don't work' (p. 58 above)). This contrasted with G's comments, which were possibly determined, in part, by his sense of what it is to be a professional linguist: `I don't use those terms ± incorrect . . . because as a linguist . . . I'm not allowed to use that sort of statement' (p. 139 above); `bank is a word which has a very wide semantic range' (p. 51 above). Such comments in no way misrepresent or falsify what native speakers of English think about the nature of the word or language. On the contrary, the responses actually foreground the bases on which such judgements are made, giving the analyst real data with which to work. Certain words were also said to be `impossible', even though they were intelligible (p. 110 above), `silly' and `stupid' (p. 111 above), `useful' [to] because N: `many sentences wouldn't make any sense at all without it' (p. 94), `complicated' M: `if you said to somebody `spell that' nobody would be able to' (p. 79). and `weird' because M: `of the unusual combination of letters' (p. 80). And the use of the labels `big', `long' and `short' was frequently ambiguous. Some applied such terms to the number of letters in the word, some to the dif®culty of pronunciation, some to the unfamiliarity and some to the `meaning'. F, for example, said of antidisestablishment that it was a word with `a lot of meaning'. Many of the words were judged to be indicators of the kind of speaker who would use them: parsimonious, for example, was said to be `pompous' and shit to be `rude' (Chapter 6). Aint was said to be `uneducated', `ignorant,' `lower class', and `lazy' (Chapter 5) and old and mature to be respectively `uncomplimentary' and `complimentary' or `diplomatic' (Chapter 6). This empirical work has thus highlighted the different possibilities of interpretation open in any communicational situation, possibilities which are a concomitant feature of human interaction, in¯uenced by, among other things, social stereotyping and the varying social, psychological and interactional histories of informants (McGregor 1986b). `Indeed, any theory of utterance interpretation must needs take these differences into account if 202

Reorientation: the Integration of Speech and Writing

the interactional and communicative functions of everyday language use are to be more fully understood' (ibid.: 180). Since what people think language is and how it works informs linguistic behaviour (pp. 33±34), this work has foregrounded the relevance that words have for the language-users themselves. This involves looking at, not an abstract concept `meaning', but rather what is `meaningful' for individual language-users. In so doing, it has supported McGregor's thesis that what eavesdroppers have to say will necessarily match their experience as actual speaker-hearers . . . that there does not have to be any de®nite list of things that must be believed by someone in order to `understand' the talk of others (1983: 277). Asking the speaker has allowed for the development of hypotheses about the kind of social assumptions that speakers must have made in order to interpret and evaluate the interview questions. As in McGregor's work, I did not attempt `to describe and account for these assumptions in any systematic way' (McGregor 1990: 101). Neither did I try to explain any variations occurring between different informants and their judgements. By stressing an individual's creativity involved in interpretation, my research has shown both a) that the idea of a language as a ®xed code shows little, if any relation to the actual behaviour of language-users, including their metalinguistic judgements, and b) the extent to which literacy does make a radical difference to the way people think about themselves and their own social activities including language. The professional linguist's talk of words, morphemes, lexemes, and the like, is just an extension of lay metalanguage. The difference is, is that segregational linguists feel the need to ®x, codify and systematise such concepts in order to explain how language works. However, as I have said earlier, `The context of the linguistic act, which includes the assumptions the speaker may hold about the hearer's intentions . . . creates the sign. The linguistic sign is not given in advance of the situation' (Davis 1990: 9). 203

Reorientation: the Integration of Speech and Writing

What any individual understands by the word word is dependent upon that individual's re¯ection on that particular word on that particular occasion. The next day, that very same individual, may well arrive at a very different understanding of that `same' word word. Although communities in western literate societies may agree that there are concepts called words, some people's understanding of words may be very different from others (as evinced in my empirical work). For some, their understanding is based upon their heavy use of dictionaries and grammars, for others the understanding may be based upon talking about, seeing or re¯ecting on how other people use these artefacts. As Harris writes, it is unlikely that such concepts as `the word' or `the English language' `arise spontaneously from individuals' untutored re¯ections on their daily linguistic intercourse' (Harris 1998: 56). Thus ordinary and extraordinary language-users, like my present informants have such a diversity of answers to metalinguistic questions precisely because of the integrational character of language. Language is embedded within everything we humanly do and this embeddedness includes our responsiveness to biomechanical, macrosocial and circumstantial factors. Because these factors will not be the same for different individuals, nor indeed for the same individual from one occasion to the next, there is no suggestion that it is possible to state which factors caused or will cause two informants otherwise quite similar (in age or occupation . . .) to offer different judgements. This study has focussed on the word word not in order to document more fully its use as a vocabulary item of contemporary English but in order, on the one hand, to examine certain features of the lay orientation typical of modern linguistic theorising and, on the other, to gain some insight into one rather central aspect of ordinary language users. These two objectives are intimately related from an integrationist perspective, which rejects any belief in innate and universal metalinguistic concepts, along with any idea of linguistics as an autonomous, intellectual activity. For the integrationist, both metalinguistic intuitions and academic linguistic theorising are culture-dependent phenomena, 204

Reorientation: the Integration of Speech and Writing

re¯ecting the concerns of individuals and societies in particular historical contexts. These two aspects of any item of metalinguistic vocabulary must therefore be studied in relation to one another if any clear understanding is to be gained of the underlying processes involved. Finally, this work has shown that asking for information is just as complex and unpredictable as the giving of it. It also has shown that ordinary language users feel under some obligation to have some opinions on the subject (see in particular the preponderance of informants' use of `as far as I'm concerned' (p. 61)). Although, in context, this phrase does not sound particularly defensive, it tends to be used when a speaker makes explicit recognition that it is part of the language game to recognise that there will be differences of opinion. Questions concerning language are similar to questions about homosexuality and religion ± i.e. questions about things and issues `everybody' is supposed to have an opinion about2. I will let two of my informants have the last word here in order to show that, while there will never be a ®nal explicit answer to that 64-thousand-dollar question `what is a word?', we can all amplify and deepen our understanding of the answers we already have. I: E: I: E: I: E: B:

What sort of words are they d'you think? What they are? Yeah Dunno ± I wouldn't even bother with `em myself personally Why? `Cause I couldn't bloody well read `em I think if somebody had said it to me I would run off to a dictionary and try to ®nd out what he was on about.

2 I am grateful to Michael Toolan p.c for this point.

205

Bibliography

Aitchison, J. (1987). Words in The Mind: An introduction to the Mental Lexicon. Oxford: Basil Blackworth Andersson, L.G. and Trudgill, P. (1990). Bad Language. Oxford: Basil Blackwell Aronoff, M. (1976). Word Formation in Generative Grammar. Cambridge, Massachussets: MIT Press Atkinson, M., Kilby, D. and Roca, I. (1988). Foundations of General Linguistics, [2nd edn.], London: Unwin Hyman Austin, J.L. (1962). How to do things with words: the William James Lectures delivered at Harvard University in 1955. Oxford: Clarendon Press Baker, G.P. and Hacker, P.M.S. (1984). Language, Sense and Nonsense: A Critical Investigation into Modern Theories of Language. Oxford: Basil Blackwell Barnes, J.A. (1990). Models and Interpretations: selected essays, Cambridge: CUP Barton, D. (1985). `Awareness of Language Units in Adults and Children' in Progress in the Psychology of Language, ed. A.W.Ellis. London: Lawrence Erlbaum Associates pp. 187±207 Bauer, L. (1988). Introducing Linguistic Morphology. Edinburgh: Edinburgh University Press Berthoud-Papandropoulou, I. (1978). `An experimental study of children's ideas about language' in The child's conception of language, ed. A. Sinclair et.al. Berlin: Springer-Verlag. pp. 55±65 Bloom®eld, L. (1935). Language. London: George Allen & Unwin Bloom®eld, L. (1944). `Secondary and Tertiary Responses to Language', Language 20, 45±55 Bolinger, D. (1946). `Visual Morphemes', Language 22, 333±339 Bolinger, D. and Sears, D.A. (1981). Aspects of Language. New York: Harcourt Brace Jovanovich Inc. Botha, R. (1968). The function of the lexicon in transformational generative grammar. The Hague: Mouton 207

Bibliography

Bowey, J.A. and Tunmer, W.E. (1984). `Word Awareness in Children' in Metalinguistic Awareness in Children, ed. W. Tunmer et al. Berlin: Springer-Verlag pp73±91 Bresnan, J. (1978). `A realistic transformational grammar' in Linguistic Theory and Psychological Reality, ed. M. Halle et al. Massachussetts: MIT Press. Pp. 1±59 Brown, K. (1984) Linguistics Today. London: Fontana Paperbacks Burgess, A. (1975) Language Made Plain, revised edn. London: Fontana Bybee, J.L. (1985). Morphology: A Study of the Relation Between Meaning And Form, Amsterdam: John Benjamins Cameron, D, (1985, 1992). Feminism and Linguistic Theory, Macmillan: Basingstoke Cameron, D. (1995). Verbal Hygiene, London: Routledge Carroll, J.B. (1972). `The Case for Ideogrpahic Writing' in Language by Ear and by Eye, ed. J Kavanagh and I. Mattingly, Cambridge, Massachussetts and London: MIT Press. pp. 103±109 Cazden, C.B. (1972). Child language and education, New York: Holt Rinehart & Winston Chafe, W.L. (1970). Meaning and the Structure of Language, Chicago: University of Chicago Press. Chomsky, N. (1957). Syntactic Structures, The Hague: Mouton Chomsky, N. (1965). Aspects of the Theory of Syntax. Cambridge, Massachussetts: MIT Press Chomsky, N. (1980). Rules and Representations. Oxford: Basil Blackwell Chomsky, N. (1988). Language and Problems of Knowledge: the Managua lectures. Cambridge, Massachussets: MIT Press Crystal, D. (1977). What is Linguistics?, London: Edward Arnold Crystal, D. (1985). Linguistics, 2nd edn. Harmonsworth: Penguin Crystal, D. (1991). A Dictionary of Linguistics and Phonetics, 3rd edn. Oxford: Basil Blackwell Culler, J. (1983). On deconstruction: theory and criticism after structuralism. London: Routledge & Kegan Paul. Davis, H.G. (1990). `Introduction' in Rede®ning Linguistics, ed. H.Davis and T. Taylor. London: Routledge Davis, H.G. (1992). `Drawing the Morphological Line' in New Departures in Linguistics, New York & London: Garland Publishing Inc. pp. 90±116 Davis, H.G. (1997). `Ordinary people's philosophy: comparing lay and professional metalinguistic Knowledge'. Language Sciences 19:1, 33±46 Davis, H.G. (1998). `What Makes Bad Language Bad?', in Integrational Linguistics: A First Reader, ed. R. Harris & G. Wolf, Oxford: Pergamon pp. 283±294 Davis, H.G. (1999). `Typography, lexicography and the development of the idea of `standard English' in Standard English: The widening debate, ed. by T. Bex and R. Watts, London: Routledge pp. 69±88 208

Bibliography

De Beaugrande, R. (1980). Text, Discourse and Process, London: Longman Downes, W. (1984) Language and Society, London: Fontana Paperbacks Fasold, W. (1990) The Sociolinguistics of Language, Oxford: Basil Blackwell Fodor, J.A., Bever, T.J. and Garrett, M.F. (eds.) (1974). The Psychology of Language, London: McGraw-Hill Fromkin, V. (1987). `The Lexicon: Evidence From Acquired Dyslexia', Language 63, no. 1, 1±22 Garnham, A. (1985). Psycholinguistics: Central Topics, London: Methuen & Co. Ltd Giddens, A. (1976). New Rules of Sociological Method: A Positive Critique of Interpretative Sociologies, London: Hutchinson and Co. Gleitman H, and Gleitman L. (1979). `Language Use and Language Judgement' in Individual Differences in language Ability and Language Behaviour, ed. Fillimore, Kempler and Wang, New York, Academic freedom Press Inc: pp. 103±126 De GoeÂs C. and Martlew, V. (1983). `Young Children's Approach to Literacy' in The Psychology of Written Language: Developmental and Educational Perspectives, ed.V. Martlew Chichester: John Wiley and Sons Goody, J.R. (ed.) (1968). Literacy in Traditional Societies, Cambridge: CUP Grootaers, W.A. (1969). `Origin and Nature of the subjective boundaries of dialects', Orbis 8, 355±384 Gumperz, J.J. (1982). Discourse strategies, Cambridge: C.U.P Gumperz, J.J. and Tannen, D. (1979). `Individual and Social Differencesin Language Use' in Individual Differences in Language Ability and Language Behaviour, ed. Fillimore, Kempler & Wang, New York: Academic Press, pp. 305±325 Hacker, P.M.S. (1986). Insight and Illusion: Themes in the Philosophy of Wittgenstein, Oxford: Clarendon Press Hakes, D.T. (in collaboration with Evans, J.S., & Tunmer, W.E.) (1980). The Development of Metalinguistic Abilities in Children, Berlin: Springer-Verlag Halle, M. `The Rules of Language' in Language Introductory Readings, eds Clark et al, New York: St Martin's Press Hamilton, M.E. and Barton, D. (1983). `Adults De®nition of `Word': The Effects of Literacy and Development', Journal of Pragmatics 7, 581±594 Harris, R. (1980), The Language Makers, London: Duckworth Harris, R. (1981). The Language Myth, London: Duckworth Harris, R. (1983). `Speech and Language' in Approaches to Language, ed. R. Harris, Oxford: Pergamon Press: pp. 1±19 209

Bibliography

Harris, R. (1986). The Origin of Writing, London: Duckworth Harris, R. (1996). Signs, Language and Communication, London: Routledge Harris, R. (1998). An Introduction to Integrational Linguistics, Oxford: Pergamon Harris, R. and Wolf, G. (1998). Integrational Linguistics: A First Reader, Oxford: Pergamon Heath, S. (1982).`What no bedtime story means: narrative skills at home and at school', Language in Society, 11, 49±76 Hoenigswald, H.M. (1964). `A Proposal For the Study of FolkLinguistics' in Sociolinguistics: Proceedings of the UCLA Sociolinguistics Conference, 1964, ed. W. Bright, The Hague: Mouton & Co. pp. 26 Howard, P. (1984). The State of the Language: English Observed, London: Harnish Hamilton Ltd. Hutton, C. (1990). Abstraction and Instance: The Type-Token Relation in Linguistic Theory, Oxford: Pergamon Press Jackendoff, R. (1993). Patterns in the Mind, Hemel Hempstead, Herts: Harvester Wheatsheaf Katz, J.J. (1966). The Philosophy of Language, New York: Harper & Row Katz, J. and Fodor, J. (1963). `The Structure of a Semantic Theory' in The Structure of Language: Readings in the Philosophy of Language, New Jersey: Prentice Hall, pp. 479±518 Kramsky, J. (1969). The Word as a Linguistic Unit, The Hague: Mouton Kreckel, M. (1981). Communicative Acts and Shared Knowledge in Natural Discourse, London: Athlone Press Labov, W. (1970). `The Study of Language in its Social Context' in Language and Social Context, ed. P Giglioli, pp. 283±307 Labov, W. (1972). `Some principles of linguistic methodology' in Language in Society 1, no. 1, 97±120 Landau, S.I. (1989). Dictionaries: The Art and Craft of Lexicography, New York: The Scribner Press Levelt, W.J.M., Sinclair, A., & Jarvella, R.J. (1978). `Causes and Functions of Linguistic Awareness in Language Acquisition: Some introductory Remarks' in The child's conception of language, ed. Sinclair, Levelt and Jarvella, Berlin: Springer-Verlag pp. 3±14 Love, N. (1998). `The Fixed-Code Theory' in Integrational Lnguistics: A First Reader, ed. by R.Harris and G. Wolf, Oxford: Pergamon pp. 49±67 Lyons, J. (1981). Language, Meaning and Context, Suffolk: Fontana Lyons, J. (1987). `Semantics' in New Horizons in Linguistics ed. J. Lyons et.al, Suffolk: Penguin, pp. 152-178 McGregor, G. (1982). `Intonation and Meaning in Conversation', Language & Communication 2, no. 2, 123±131 210

Bibliography

McGregor, G. (1983). `Listeners' comments on conversation', Language & Communication 3, no, 3, 271±304 McGregor, G. (1984). `Conversation and Communication', Language & Communication 4, no. 1, 71±83 McGregor, G. (1985). `Utterance Interpretation and the Role of the Analyst', Language and Speech 28, 1-28 McGregor, G. (1986a). Language for Hearers (ed.), Oxford: Pergamon Press McGregor, G. (1986b). `The hearer as ªlistener judgeº: an interpretive sociolinguistic approach to utterance interpretation' in Language for Hearers, ed. McGregor 1986a, Oxford: Pergamon Press, pp. 153±184 McGregor, G. (1986c). `Listening Outside the Participation Framework' in The Art of Listening, ed. By G.McGregor and R. White 1986, Kent: Croom Helm Ltd. pp. 55±72 McGregor, G. (1990). `Eavesdropper response and the analysis of everyday communicative events' in Reception and Response: Hearer Creativity and the Analysis of Spoken and Written Texts, ed. by G. McGregor and R. White, London: Routledge. pp. 99±111 McGregor, G. and White, R.S. (eds) (1986). The Art of Listening, Kent: Croom Helm Ltd. McGregor, G. and White, R.S. (eds) (1990). Reception and Response: Hearer Creativity and the Analysis of Spoken and Written Texts, London: Routledge Michaels, L and Ricks, C. (eds.) (1980). The State of the Language, California:University of California Press Miller, G.A. (1978). `Semantic Relations among Words' in Linguistic Theory and Psychological Reality, ed. by M. Halle, J. Bresnan, and G.A. Miller, Massachussets, MIT Press. pp. 60±118 Miller, G.A. (1981). Language and Speech, San Francisco: W.H. Freeman and Company Montagu, A. (1968). The Anatomy of Swearing, London: Rapp and Whiting O'Grady, W. (1996). `Syntax: the analysis of sentence structure' in W. O'Grady et al, Contemporary Linguistics: An Introduction, London & New York: Longman pp. 181±244 O'Grady, W, Dobrovolsky, M. & Aronoff, M. (1989). Contemporary Linguistics: An Introduction, New York: St Martin's Press Olson, D. (1984). `ªSee! Jumping!º Some oral language antecedents of literacy' in Language and Literacy from an educational perspective, ed. by N. Mercer, Milton Keynes: OUP. pp. 221±230 Olson, D. (1991). `Literacy as metalinguistic activity' in Literacy and Orality, ed. by D. Olson and N. Torrance, Cambridge: CUP. pp. 251±270 Olson, D.R., and Torrance, N. (eds.) (1991). Literacy and Orality, Cambridge: CUP 211

Bibliography

Olson, D.R., Torrance., Hildyard, A. (eds) (1985). Literacy, Language and Learning. The nature and consequences of reading and writing, Cambridge: CUP Ong, W.J. (1982). Orality and Literacy: The Technologizing of the Word, London and New York: Methuen Palmer, F. (1984). Grammar, Harmondsworth: Penguin Papandropoulou, I. And Sinclair, H. (1974),`What is a word? Experimental study of children's ideas of grammar', Human Development 17, 241±258 Passy, P. (1932). Les Sons Du FrancËais, Paris: Henri Didier Pateman, T. (1987). Language in Mind and Language in Society: Studies in Linguistic Reproduction, Oxford: Clarendon Press Piaget, J. (1929). The Child's Conception of the World, London: Routledge and Kegan Paul Quine, W.V.O. (1960). Word and Object, Cambridge, Mass: MIT Press Read,, C. (1975). Children's categorizations os speech sounds in English, National Council of Teachers of English: Illinois: Urbana Reddy, M.J. (1979). `The Conduit Metaphor ± A Case of Frame Con¯ict in Our Language about Language' in Metaphor and Thought, ed. by A. Ortony, Cambridge: CUP. pp. 284±310 Ricks, C. (1980). `Prefatory Note' in The State of the Language, Berkely, Los Angeles, California: University of California Press, pp. xi±xii Ricks, C. and Michaels, L. (1990). The State of the Language: (1990) Edition), London: Faber and Faber Risch, B. (1987),`Women's derogatory terms for men: That's right, ªdirtyº words', Language in Society, 16, No. 3, 353±358 Ross, J.R. (1979). `Where's English' in Individual Differences in Language Ability and Language Behaviour ed. by C. Fillimore, D. Kempler and W. Wang, W. S-Y, New York: Academic Press pp. 127±47 Sapir, E. (1921), Language, An Introduction to the Study of Speech, London: Harcourt, Brace & World, Inc. Sapir, E. (1949). `The Psychological Reality of Phonemes' in Selected Writings in Language, Culture and Personality, ed. D. G. Mandelbaum, Berkeley, Los Angeles, pp. 46±82 Saussure, F. de (1922). Cours de linguistique geÂneÂrale, trans. Roy Harris, London: Duckworth Taylor, T. J. (1981). `A Wittgensteinian Perspective in Linguistics', Language & Communication, 1, no.2/3, 263±274 Taylor, T.J. (1990). `Normativity and Linguistic Form' in Rede®ning Linguistics, ed. by H.G. Davis and T. J. Taylor, London: Routledge, pp. 118±148 Taylor, T.J. (1992). Mutual Misunderstanding Durham: Duke University Press Toolan, M. (1996). Total Speech: an integrational linguistic approach to language, Durham and London: Duke University Press 212

Bibliography

Torrance, N. and Olson, D.R. (1987). `Development of the metalanguage and the acquisition of literacy', Interchange 18, 136±146 Trudgill, P. (1983/1995) , Sociolinguistics: An Introduction to Language and Society, Middlesex: Penguin Books Tunmer, W.E., Pratt, C., and Herriman, M.L. (1984). Metalinguistic Awareness in Children, Berlin: Springer-Verlag Wittgenstein, L. (1958). Philosophical Investigations, trans. G.E.M. Anscombe (2nd edn), Oxford: Basil Blackwell Wittgenstein, L. (1975). Philosophical Remarks, ed. by R. Rhees, trans. R. Hargreaves and R. White, Oxford: Basil Blackwell Wolf, G. (1992) New Departures in Linguistics, New York & London: Garland Publishing.

213

Index

Aitchison, J. 12, 17, alphabet, 13, 69±70 Amis, K. 190 Andersson and Trudgill, 150, 152 Aristotle 107 ask-the-speaker, ix, 2, 31f., 36, 203 Atkinson et al., 98f. Austin, J.L. 107n bad language, 136, 148ff. Baker and Hacker, 101f, 116 Barnes, J.A. 35 Barton, D. 56f., 62, 68, 81, 196, Bauer, L. 9±10, 16, 54±5 Berthoud-Papandropoulou, I. 145, 196 biomechanical 30, 193, 204 biplanar, 193 Blakemore, 12n Bloom®eld, L. 6n, 32f., 87±8n, 107n Bolinger, D. 82ff. Bolinger and Sears, 192n Botha, R. 11 Bowey & Tunmer, 17 Branson, Richard, 148±9 Brown, K. 4, 117 Burgess, A. 56 Cameron, D. 138, 190

Cazden, C. B. 196 Chafe, W. 21 Chomsky, N. 14f, 37, 88n, 117, 196 circumstantial 30, 204, code theory, 21f. cognitivist, 13 conduit metaphor, 24, 28 Clinton, 24±5 consonants 44, 70, 76, 77, 78, 80, 81 Crystal, D. 7, 10n, 190 Davidson, D. 31 Davis, D. ix Davis, H.G. 6, 11, 13, 115n 148, 161, 200, 203 de Beaugrande, R. 201 de Goes & Martlew, 18 de®nition, 43, 95f., 98ff. descriptions, 187f. determinacy, 5, 6, 11, 21f., 66 diachronic, 51 diacritics, 68, 199 dictionary, 24ff., 43, 50, 63, 98ff., 116, 137, 153, 155f., 200 Downes, W. 37 Dummett, M. 31 ethnomethodology 36f. etymology, 50±51

215

Index

IAISLC ix ideal-speaker-hearer, 196 indeterminacy, ix, 1, 22, 36, 197 informant pro®les 44ff. integrational ix. 29ff., 36f., 191ff. intuitions, 6, 9, 16, 204

exclamation mark, 156, 200 family resemblances, 27 Fasold, R.W. 32 ®xed code, 24, 31 Fodor, Bever and Garrett, 14 folk-theoretical, 1, 147ff. Frege, G. 31 Fromkin, V. 12

Jackendoff, R. 3, 4

Garnham, A. 56 generativist, 31 Giddens, A. 35 Gleitman and Gleitman, 33 Goody, J. 195 O'Grady, W. 117f. O'Grady, Dobrovolsky and Aronoff, 56n grammar, x, 12, 84, 112, 117ff. grammatical, 42, 75, 146, 156, 160, 201 grammatical error, 136, 138. grammatically incorrect, 138±143 Grice, H.P. 115 Grootaers,W. 34 Gumperz, J. 35, 37±38 Gumperz and Tannen, 35f., 38 Hacker, P. 103, 104ff., 105±6, 115n Hakes, D.T. 196 Halle, M. 9 Halliday, M. 19n Hamilton and Barton, 119, 199 Harris, R, ix, x, 5n, 10n, 19n, 22±3, 29f., 70, 115n, 118, 192, 193, 204 Harris, Z, 117 Heath, S. 194 historical linguistics, 50±1 Hoenigswald, H.M. 34 homonymy, 67n, 82n Howard, P. 68, 150 Humboldt, 6n Hutton C, ix, 23, 67

Katz, J.J. 8 Katz and Fodor, 101n Kramsky, J. 8 Kreckel, M. 35f. Labov, W. 33, 37, 135, 146 Landau, S. 100 langue, 7n language game 26n, 192, 205 lay speakers, ix±x, 1, 2n, 23±4, 31f., 34±44, 147, 173ff. Leibnitz, 6n letters, 70ff. Levelt, Sinclair and Jarvella, 196 lexeme, 54, 67n, 86, 203 lexicographic, 64, 86, 98f. linguistic form, x, 6, 43f., 49ff., 86, 87, 118 linguistic meaning, x, 6, 87ff., 118 linguistic sign, 5, 7 literacy, 56, 63, 118f, 156, 191ff. literate, 145, 192ff. Locke, 6n, 132 Love, N, ix, x, 23, 33±34, 115n Lyons, 10, 55, 67n, 87, 91f, 101, 115 macrosocial 30, 204 McGregor, G. 35f, 51±2, 193, 202±3 McGregor & White, 35f. meaning, 43, 50, 52, 55, 87ff., 98 mental lexicon, 11±12, 15±16, 17, 28 metalanguage, 193, 203, 204 metalinguistic, ix, x, 2, 17, 23, 24, 31, 33±34, 36, 98, 114, 199, 203±4

216

Index

metalinguistic awareness, 18, 56, 196, 201 metaphor 24, 42, 103, 107ff. Michaels and Ricks, 150 Miller, G. A. 11±12 Montague, A. 150 morpheme, 66, 83, 86, 203

reading, 69, 71, 72, 85, 86 Reddy, 24, 28 Ricks, C. 100 Ricks and Michaels, 150, Risch, B. 151±2 Ross, J.R. 145 Russell, B. 31

national curriculum, 2n normativity, 28, 190

Sapir, 9, 13, 14, 17, 54, 69, 82, 134 Saussure, 3ff., 21, 37, 49n, 68±70, 69n, 98 , 194f. schooling, 118, 119, 139, 143, 199 science, 2, 5, 6, 19, 21 segregational 29, 31, 193, 198, 203 semantic 49, 63, 68, 86, 97 semi-literate, 57 slang, 63, 83, 137, 149, 153f. social science, 1, 31, 34±35 sociolinguistics, 35, 36, 134f. sociolinguistic variable, 135 sounds 70, 76, 77, 83, 86, 184 speech circuit, 3, 6, 12 spelling, 43, 49, 66, 69, 70, 71ff., 82, 84, 85, 86, 98 swearing, 148 swearword, 63f., 148ff. synchronic 51 synonyms 51

OED2, 69, 107, 153±4 Olson, D. 192, 193, 194 Olson and Astington, 200 Olson & Torrance, 192 Olson, Torrance and Hildyard, 192 orthographic, 49, 50, 55, 63, 70, 86 orthography 43, 49, 50±1, 55, 63, 67 Ong, W. 53 ostensive de®nition, 104n Palmer F. 117 Papandropolou and Sinclair, 196 parole, 7 Partridge, E. 150 parts of speech, x, 11, 42, 117ff. Passy, 10f. Pateman, T. 22±3 pedagogy, 55, 73 Piaget, J. 17±18, 196 polysemy, 67n Port Royal, 132 pragmatics, 12n pre-literate, 16, 193 principle of compositionality, 98, 100ff. pronunciation, 50, 69, 79 psychological reality, 6, 7, 9, 14, 15, 17, 68 Quine, W. 8 Read, C. 81

taxonomic linguistics, 6 Taylor, D.M. 112 Taylor, T.J. ix±x, 2, 21f., 28f. telementation, 5, 68 Tooke, 6n Toolan, M, ix±x, 19n, 205n Torrance and Olson, 200 transformational generative grammar, 14f. Trudgill, P. 147, 150±2 Tunmer, Pratt and Herrimann, 145, 196 type-token, 23, 67n, 118

217

Index

ungrammatical, 145 visual morphemes, 83, 86 vowels, 44 Weiner, 190 Wells, C.J. x Wittgenstein, 1, 26ff., 103, 104, 104n, 106, 116, 133 Wolf, G, ix

Words: concept 104f. function 91ff. legal and religious words 165ff. old words, 163ff. scienti®c words, 170ff. spoken 178f., 193 written 167, 169, 176ff., 193 writing, 5, 7, 9, 13, 23, 28, 39, 69, 77±8, 137, 191ff. writing systems, 9

218